South Ronaldsay, Howe Of Hoxa

Four drawings by George Petrie of sites in Orkney in sketchbook no. 6 (MS 28/487/8) in the Society of Antiquaries of Scotland Collection.

A broch occupies a natural ridge to the centre of a narrow isthmus. This elevated site offers good views over the surrounding landscape and is clearly visible from as far away as St. Mary's Holm on Mainland. The broch was partially excavated by Petrie in 1848 and has been largely 'renovated' since then. The interior was divided by radial slab divisions, although, those now in evidence may be reconstructed rather than original. The internal ground surface has been levelled out and roughly surfaced. The inner wall face has been rebuilt with further, mortar-bonded courses, added. The outer wall face, which is not now visible, is said to have been 14' thick and to have stood to 8' high. The exterior appears not to have been fully exposed; substantial deposits survive around the circuit of the broch. Topographic anomalies indicate the presence of further structures, possibly houses, built over accumulated debris to the immediate E of the broch. Further to the S end of the ridge, two uneven areas may represent the remains of yet more settlement.

Moore and Wilson, 1997

Coastal Zone Assessment Survey

ND49 5 HOWE OF HOXA (‘Brough of Hoxay’)

ND/42529396 (visited 16/7/63 and 13/7/1985)

This probable solid-based broch stands on the north shore of South Ronaldsay I., on the summit of a hog-backed ridge (the “howe” in the name) among flat ground and overlooking the beach.

Principal Gordon visited the site towards the end of the 18th century and reported seeing a curved gallery on top of the mound [3]. This site seems to have been only the second Orkney broch to have been dug into for antiquarian reasons -- the first being Burgar in 1825 (HY32 1), and the project was carried out by George Petrie in 1848 [6, 8, 9]. However the site was almost immediately afterwards mutilated with mortared masonry by the well intentioned efforts at conservation carried out by the proprietor [6, 108 ff.]. Although several features are now hidden or destroyed because of this, the cementing has nevertheless ensured that -- in contrast to Burray East for example (ND49 1) -- much of the structure is still visible and the interior is relatively free of debris. However the almost complete lack of understanding of broch architecture at the time the work was done means that the existence of potentially important features like the intra-mural stair were unsuspected and unsought while others were covered up. A plan prepared at the time has been published [10, fig. 3.23).

Apart from the area around the main entrance, only the interior of the tower was exposed and the site now looks like a knoll containing a large round hole lined with masonry; the outer face seems to have been further buried by material thrown out from the interior. The upper part of the inner face of the wall, to about half way down, was pointed with cement; indeed much of the top of the wall seems to have been completely reconstructed, as can be seen on the north side where masonry has been added directly on top of the original shelf scarcement, producing a “reverse scarcement” effect. This modern superstructure is clearly indicated above a 'ledge' on the elevation above the plan in Petrie’s notebook [10, fig. 3.23].

Some modern features have also been added, like a stone platform of flags which once had a flagpole on it and which stands beside the modern entrance. This doorway faces east and is completely modern, and one cannot tell now whether any of the original passage remain. This was exposed in 1848 and appears to have been of the standard broch design, with one set of door-checks and probably a bar-hole leading in to a mural cell immediately to the right (but with its door leading to the interior of the broch) [6, 109, fig. 121]. The present entrance is short, cemented and lacks any door-checks; no guard cell can now be seen.

The interior diameter is 9.05 m (29 ft. 8 in.) and a secondary wall lining has been built round at least part of the inner court, varying in thickness from 30-90 cm (1-3 ft.). This rises only 2.29 m (7.5 ft.) above the present floor and at its highest point ends just below a shelf scarcement 13 cm (5 in.) wide made of projecting slabs. At one point on the north side this scarcement can be seen above the secondary wall and emerging from the cemented masonry; the conservers evidently had no idea of the significance of the ledge. Opposite the doorway the secondary wall blends with the cemented primary face, indicating that extensive reconstruction has taken place. Radial slabs are still visible set into the secondary wall, and seem to have been somewhat larger when first exposed [6, fig. 120]. They evidently divided the floor of the interior up into at least six radial partitions, rather like a wheelhouse. The drawing also shows rotary querns on the floor [10, fig. 3.24].

The building almost certainly possesses an intra-mural stairway leading to upper galleries in a hollow wall. At about 9 o’clock two upright slabs are set into the inner wall face which, even though no sill stone seems to be preserved, must surely be the jambs of a raised door, now blocked up; this, judging from similar doorways in other Orkney brochs, probably leads to the stair [10, fig. 3.23]. The feature is recorded in a drawing made soon after the excavation [6, fig. 120], but that shows the masonry between the jambs as neat; in fact it is now a very crude rubble blocking and might have been inserted by the proprietor soon after the excavation. Other shelf-like structures in the inner wall face shown in the same drawing are no longer visible.

It should be noted that the only authentic account of the broch as it was found when excavated is that by Petrie. The other descriptions were second-hand and some of the statements in them describe features seen by people who had seen the ruins only after they had been "conserved". Even the mural cell seems not to have been seen by Petrie since he says "There appears to have been a chamber in the thickness of the wall ... " [8, 23] and marks this at 9 o’clock on his sketch plan [10, fig. 3.23]. It is not clear whether the drawing reproduced by Wilson [6, fig. 120] was made before or after conservation.

Few finds were recorded [8], mostly of stone, and they include some querns, evidently rotary [6, fig. 120], a mortar and pestle and 2 small cups or lamps (one apparently a steatite handled lamp [10]), both 11 cm (4.5 in.) deep and 13 cm (5 in.) and 18 cm (7 in.) in diameter respectively. There were also saddle querns and rubbers [5, 122].

Dimensions: external diam. c. 17.68 m, internal 9.14 m; the wall thickness is about 4.27 m [10] so the wall proportion would be about 48%. In 1985 the author made a plan of the interior wall face, below the reconstructed parts, with a steel tape and theodolite; the result showed that the structure had been built very exactly around a true circle with a radius of 4.53 +/- 0.06 m, or a diameter of 9.06 m (29.70 ft.).

Sources: 1. OS card ND 49 SW1: 2. Low 1879, 22: 3. Gordon 1792, 257: 4. Petrie 1872, 361-62: 5. Thomas 1852, 119-22: 6. Wilson 1863, 2, 107-10: 7. Tudor 1883, 335: 8. Petrie 1927, 23: 9. RCAHMS 1946, 2, no. 815, 283-4: 10. Hedges et al. 1987, 102-04.

E W MacKie 2002