Dun Alascaig

Broch, Dun Alascaig.

The remains of this broch, which was described while it was still standing to a considerable height (Maitland 1757, 145; Cordiner 1788, 118) can still be seen on the S side of and about 30 ft above the Kincardine-Edderton road about half way between the 9th and 10th milestone from Tain. The site is just under 100 ft above sea-level. All that now survives of the structure is a row of five massive stones representing the footings of the outer wall face on the W sector, together with a few others elsewhere on the periphery and a single small showing of the inner wall-face. From these the original diameter of the broch may be estimated at about 58 ft; the wall-thickness measurable at one point as 12 ft 3 in, the internal diameter may be put at about 34 ft. The site need not be revisited.

Visited by RCAHMS 9 September 1943

OS 6" map, Ross-shire, 2nd ed., (1907)

Dun Alascaig NH 656 868 NH68NE 11

Little now survives of this broch, which was still standing to a height of 4.5m in the 18th century.

RCAHMS 1979, visited November 1977

(Maitland 1757, i, 145; Pococke 1887, 111-3; Young 1962, 186)

NH68 1 DUN ALISAIG ('Dun Alascaig', 'Dun Alliscaig'*, 'Dunaliscaigh', 'Dun Agglesag', 'Doniskaig')

NH/6569 8682 (visited in 1989 and 25/7/03)

Site of a broch, probably solid-based, in Edderton, Easter Ross; it stood just above the road on the rising ground on the south shore of the Dornoch Firth, one of the very few on that side of the estuary. It is now a total wreck and only a heap of stones and about twenty earth-fast boulders can be seen [9], though another recent report says that a single course of the outer face survives almost intact around the north and west arcs [1], and some years ago Alison Young found the triangular lintel of the main entrance lying overgrown among the grass. This stone was exceptionally large, the base being 1.5m (5 ft) long and the sides 85cm (2ft 10 in) and 1.05m (3ft 6 in) [9]. The author did not locate this site until his third visit and even then was unable to find the triangular lintel or the basal course of the outer face. He saw nothing but a scatter of rubble among the trees (it is impossible to find a car-parking space close to the site).

Historical background

Dun Alisaig is of particular interest because of the contrast between its one- time fame and its present totally destroyed state. In the later 18th century it was one of four well-preserved brochs of which detailed descriptions had been published and which were therefore well known to the interested antiquaries of the day. However the three others – the two fine examples near Glenelg on the west coast (Dun Troddan and Dun Telve – NG81 3 and NG81 2) – and Dun Dornadilla in Sutherland (NC44 2) – were much more remote and difficult of access. Dun Alisaig, being close to and just above the main road a few miles west of Tain, was much more easily accessible, as the number of 18th-century visits testifies.

The brochs near Glenelg had been described in the early 1720s by Gordon (1726), and again about fifty years later (and after they had been substantially damaged) by Pennant, but thereafter they remained more or less ignored until Curle's early 20th century excavations. Dun Dornadilla was first described in print by Bishop Richard Pococke in 1760.

Dun Alisaig by contrast may have been known as early as the 1520s. The Scottish historian Hector Boece, after describing the town of Tain, wrote (in Latin) –

"There are preserved, in a certain valley in Ross, two edifices of antiquity, monuments of a round shape, made in the form of bells." [7].

This suggests that in 1526 both Dun Alisaig and another unknown broch nearby closely resembled Mousa in their profiles [2]. The second site may be Carn Mor at Birchfield some miles further up the Kyle of Sutherland (which must be the “valley in Ross” – NH49 1). Indeed if the belief is correct that Dun Alisaig is depicted on Sueno's stone, about 40km to the south-east as the crow flies [10], then this would be independent evidence that as late as the 9th century Dun Alisaig (or the other one in Ross-shire) was still a tall tower with a single doorway. If the doorway depicted on the stone is about 1.8m high the height of the tower as shown can be estimated. However some believe that the doorway represents a round-arched rather than a lintelled form which would point to an early medieval round tower rather than a broch. To the author – naturally – the roundness looks more like erosion.

In the last two decades of the 16th century Timothy Pont was making his maps of Scotland which, less than a century later, were incorporated into Blaeu's Atlas (Blaeu 1967). Colour versions have recently been printed (Stone 1991). The map of 'Southerlandia' shows a small circle marked 'Dun Alliscaig' on the south shore of the Dornoch Firth.

The broch was described in detail by the historian William Maitland in 1757, who attempted to infer the likely purpose of such “Danish forts” from the archi-tectural details he saw there. In the following years this well-preserved broch was visited and described – often accompanied by a ground plan with or without an elevation – by Bishop Richard Pococke in 1760 (Kemp 1887), by Charles Cordiner (1776) and by James Anderson a few years later (1779). This is the last published account of the upstanding broch that has been traced here.

Until recently the author thought that the Danish archaeologist J J Worsaae saw the ruined tower – in 1846 during his tour of Scotland – still standing about 20 feet high, and sketched it in his notebook (MacKie 1995). However comparing Worsaae’s drawings with those published by Charles Cordiner it is quite clear that the former were copied from the latter (which was doubtless in the library of Dun-robin Castle, where he stayed) and that Worsaae probably did not see Dun Alisaig himself. Although Cordiner’s elevation is quite fanciful, and shows a series of what look like large voids with shallow, arched roofs facing into the interior, Worsaae reproduced it despite having sketched the Backies broch during its excavation and thus being familiar with the basics of this kind of Iron Age architecture.

During all these decades Dun Alisaig was regarded by antiquaries as the typical 'Danish fort' to cite and discuss. The 18th century descriptions are reproduced in Appendix 1.

Some time after Cordiner’s visit the broch was totally destroyed, probably by road builders, and memory of it sank into oblivion [12] (Watson says that it was destroyed at about 1818 [8]). The even better preserved broch of Mousa in Shetland might have supplanted Dun Alisaig in reputation except for its remoteness; it is marked on Blaeu's Atlas as 'the ancient burgh of Mousa' so Pont knew about it, perhaps saw it, in the late 17th century. The earliest detailed description of it was by George Low in 1777 but this remained unpublished until 1879 (HU42 6). The 19th century observers and excavators looked elsewhere – to Orkney, Caithness and Sutherland and to the Western Isles – for examples to describe and, later, to excavate [Vol. 1, 31ff).

Early descriptions

William Maitland visited the site in about 1755 and wrote a detailed description, though without a drawing (Appendix 1). He found it standing about 6.1m (20 ft) high, but with large quantities of rubble lying round about which implied that it was originally somewhat higher [3]. He gives the overall diameter as 16.7m (54ft 9 in) which is not much more than that of Mousa, but the wall, at only 3.66m (12 ft) thick, was thinner. He was probably measuring above the surrounding rubble which may well have accumulated to a depth of 1.8m at least, so the wall at ground-level was doubtless wider.

The single entrance passage was about 1.83m (6 ft) high and 97cm (3ft 2 in) wide, with a massive triangular outermost lintel. Thus any rubble which had accumulated at the base of the outside of the tower evidently did not obstruct the doorway. The description suggests that there was a doorway in each side of the entrance passage – presumably from the door-frame (which is not mentioned) – "leading into the cavity between the walls, which extends round the inside of the building, of the breadth of five feet and ten inches." This sounds at least partly like the basal gallery of a ground-galleried broch. He also describes the wall as standing up to three storeys, the lowest being 2.14m (7 ft) high with two doorways (presumably four if there were in fact four cells), the middle being 2.44m (8 ft) high (very lofty for a first-floor gallery) and with four 'windows', and the upper probably of the same height but dilapid-ated at the top.

Bishop Pococke saw the broch in June 1760 [4] and gives the height as 4.58m (15 ft); he also provides a useful ground plan. From this it is clear that Dun Alisaig had the usual door-frame in the main entrance passage, 1.22m (4 ft) from the outside. The intramural openings on either side can now be seen to be a pair of elongated guard cells each of which apparently communicates with the intramural gallery beyond (but Cordiner’s drawing contradicts this). He also shows three doors from the central court (the diameter of which is 9.15m or 30 ft) to the gallery at 9, 12 and 3 o'clock.

Charles Cordiner visited the site in about 1775 but his description is less clear [4]. He gives a diagrammatic ground plan which shows four mural chambers with oval ends, in contrast to Pococke's. Unfortunately it is impossible now to disentangle from these conflicting and imprecise accounts the true nature of the ground-level storey of Dun Alisaig. The fact that these Level 1 features were almost certainly completely buried under fallen rubble helps to explain the different interpretations

James Anderson saw the broch in the summer of 1775 and, even though he lost his notes and had to rely on his memory, his account is still useful [6, 248-55 and pl. XXII]. His sketch plan and elevation suggest that the broch was at least 6.1m (20 ft) high when he saw it, even allowing for the exaggeration of the vertical scale [6, 255]. He may have been allowing for about 1.8m of the wall footings being concealed by rubble, as Pope did in his drawing of Dun Dornadilla. He mentions that about twenty years earlier – that is presumably just after Maitland's visit – the local laird had removed about 2.44 - 3.05m (8-10 ft) of stonework from the top of the wall to use for building purposes [6, 250].

Anderson described the entrance with its massive triangular outer lintel covering a passage which was about 1.22m (4 ft) wide and partly blocked with debris. He gave the sides of the lintel as 6 ft and the thickness as 4ft., but these measurements are exaggerated [9]. He could not find the longitudinal gallery inside the wall which had been described by previous observers but he did locate and climb the intramural stair; presumably he did not look behind him as he climbed or he would surely have seen the galleries (perhaps it was too dark). His description of a continuous stair roof composed of stepped lintels rising parallel to the steps of the stairway sounds somewhat improbable. When considering the descriptions of 18th century observers one must make allowances for the absence at that time of a real understanding of the detailed features of broch architecture.

The central court must have been full of at least 1.8m of rubble at that time and the interior doorway by which he got access to the stair (presumably that at 9 o'clock) may in fact have been the large void which presumably existed over the ground-level door. In this case the bottom part of the stair would have been concealed under rubble, and indeed he describes some lower steps as being partly covered with rubbish. He doubtless walked over this rubble, to the right of the opening, to reach the exposed steps.

The stairs evidently ascended by way of a series of landings about 1.83m (6 ft) long, each of which was presumably formed by the lintels of the wall gallery below; the first landing should have been on top of the presumably solid wall base, or on top of the presumably lintelled roof of one of the cells. Anderson mentions one complete flight with such a landing at either end; the flight below the lower landing was partly blocked by rubbish (confirming that he was at first-floor level) and that above the upper one, in “an imperfect state”, emerged on to the wallhead as it existed at that time. If each of the two landings was at the base and roof of the first-floor gallery respectively, the broch was probably between 4.58m and 5.49m (15-18 ft) high in 1775. It must have had one intact upper gallery and a second almost intact if not completely so. A third complete gallery doubtless existed until about 1735. Anderson describes voids in the inner wallface and these are shown in his elevation, albeit in a somewhat random manner.

Anderson specifies that the roofs of all these “rooms” or cells inside the wall were lintelled rather than corbelled. His plan shows a long intramural stair rising clockwise from an internal doorway at about 9 o'clock, and his elevation shows the triangular lintel clearly, with the stair emerging at the wallhead at about 12 o'clock.

Structural analysis

It seems most probable that Dun Alisaig was technically a solid-based broch, albeit one with a large number of mural cells in Level 1. Pococke's drawing of each of the guard cells connecting with a mural gallery behind seems to be mistaken; Cordiner’s sketch clearly shows them to be isolated but not far from the ends of the dumb-bell-shaped mural chambers beyond. Although Cordiner was not able to trace the stairway – so that he shows a dumb-bell chamber at 9 o'clock instead of the usual guard cell at the foot of the stair – his sketch confirms the separateness of all the cells.

Thus, including the guard cells, there seem to have been no less than eight chambers scattered round inside the wall in Level 1 all of which, it seems, had lintelled roofs rather than corbelled domes. On the other hand it is just possible that these lintelled roofs imply that the early observers were seeing sections of a continuous ground-level gallery rather than separate chambers, and that Pococke's impression was right. However this interpretation seems less likely, not least because no other such Hebridean broch forms are known in the area. There can be no certainty about some of the architectural features of this site.

The first continuous intramural gallery was therefore probably in Level 2 and was interrupted, as usual, by the stair rising from ground-level at 9 o'clock. The stair evidently had at least two landings in the surviving part which suggests (if it started at about 9 o'clock) that it led to a large raised doorway out onto a fairly low scarcement ledge (not mentioned by any of the observers) at between about 10-11 o'clock. Cordiner’s rather strange sketch suggests that there were a number of voids visible in the interior wallface at the time of his visit, including one which fronted the chamber over the entrance passage. There will have been at least two more galleries – Levels 3 and 4 – surviving until early in the 18th century.

Dimensions (from Maitland [3]): internal diameter 9.15m (30 ft), wall base (battered externally) 3.76m (12ft 4 in), external diameter 16.70m (54ft 9 in). The wall proportion is thus about 45%.

Sources: 1. NMRS site no. NH 68 NE 11: 2. Boece 1520: 3. Maitland 1757, vol. 1, 145: 4. Pococke 1760 (Kemp, ed. 1887): 5. Cordiner 1776, 118 and pl. XX: 6. James Anderson 1779, 241: 7. Joseph Anderson 1870, 192-3: 8. Watson 1904, 31: 9. Young 1962, 186: 10. MacKie 1975a, 204: 11. RCAHMS 1979, 23, no. 185: 12. MacKie 1995a, 144.

E W MacKie 2007