Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset

South Ronaldsay, Windwick, Cairns O' The Bu

Broch (Iron Age), Midden(S) (Period Unassigned), Pit (Period Unassigned), Roundhouse (Prehistoric), Settlement (Iron Age), Souterrain (Prehistoric)(Possible), Structure(S) (Period Unassigned), Animal Remains(S), Carved Object, Mount(S) (Antler), Rotary Quern, Spindle Whorl(S)

Ordnance Survey licence number 100057073. All rights reserved.
Canmore Disclaimer. © Copyright and database right 2021.

Toggle Aerial | View on large map

Digital Images

Administrative Areas

  • Council Orkney Islands
  • Parish South Ronaldsay
  • Former Region Orkney Islands Area
  • Former District Orkney
  • Former County Orkney

Archaeology Notes

ND48NE 14 4542 8688

See also ND48NE 9.

Although it is known by some as "Cairns of Flaws" the older inhabitants of Windwick know a rise in a field at ND 4542 8688 by its old name of "Cairns o' the Bu" (see ND48NE 9).

Immediately to the E is a wall 110.0m long which was built with stones from several "Picts Houses" discovered over the years in the vicinity of the rise, and in the area to the S. Many large lintel stones were unearthed, and stones are still being turned up by the plough. Large quanatities of bones, limpet and whelk shells, and a broken deer antler have been found (Mr John Gray, Hestly). It is believed that a

structure still remains intact under the main rise (Mr J Halcro, 6 Erland Terrace, St Margaret's Hope). This sounds like an extensive domestic site, probably similar to ND48NE 9.

This is RCAHMS (1946) first mound.

Visited by OS (IMT) 28 April 1973.


Excavation (10 July 2006 - 30 July 2006)

ND 4542 8688 Excavation was undertaken at the Cairns, Windwick Bay, from 10–30 July 2006 as part of the contributor’s research investigating Orcadian souterrains and their contexts. The site comprises a low mound with an overall extent of some 40m, which itself rests upon a larger, presumed natural, rise. A trench 15 x 15m in area was opened over the centre of the mound in order to confirm the presence of a souterrain described in 1903.

Massive roundhouse. Evidence for a very substantial, thickwalled roundhouse was uncovered. This structure has an internal diameter of 11.8m and is estimated to possess a total diameter of c22m. The wall of the structure is over 5m thick and the height is over 1m in the few areas tested. Although no definite traces of intramural galleries or staircases were detected it remains possible that such features remain to be discovered and that this very substantial building represents a broch-type structure. A gap in the masonry on the SW of the roundhouse may be an entrance but could equally be an opening into an intramural chamber. Finds included many sherds of pottery, stone tools and animal bones from the upper rubble layers and the wall heads. Of particular note was the discovery of a cache or hoard of 12 long-handled ‘weaving-combs’ present in a burnt deposit resting upon the deliberately reduced wall head of the roundhouse on the eastern side. Half of these combs are beautifully decorated with saltires and other geometric designs. The combs appear to have been part of a depositional event that occurred at the end of, at least one period of, the use of the structure, perhaps as part of a decommissioning episode.

Sunken feature/possible souterrain. In the centre of this roundhouse a rectangular chamber was partially revealed. This is almost certainly the feature briefly uncovered in the early 1900s. The removal of early 20th-century backfill revealed that the excavated portion was rectangular in shape, 2.6m long and 1.6m wide, and oriented approximately NW to SE. On the S the structure is furnished with a pair of parallel upright partitions,

or stalls, that partly rest against the wall faces but are not integrated in the stonework. These indicate that the chamber continues beyond to the S. On the N of the chamber three large orthostats placed end to end appear to form a screen and this may be the end of the chamber here or a later blocking off of a more extensive structure beyond. On the W side coursed side-walling gives way to an area of rubble 0.8m wide framed by two uprights, which may well represent a doorway framing an entrance passage leading into the chamber and blocked in antiquity.

Relatively high up in the disturbed soils within the sunken feature several large cracked slabs tipping into the chamber, and capped with clay, are likely to be remains of the roofing arrangements. Finds from the fills within the possible souterrain chamber included a saddle quern and three upper stones for saddle quern grinding; animal bone and shell. Initial work indicates that this subterranean chamber is stratigraphically later than the roundhouse, although further planned work will be required to confirm this.

Later activity inside the remains of the roundhouse. The interior of the massive roundhouse is choked with rubble. On the NW this rubble has been dug into, apparently during the late Iron Age, and a very well laid out set of stalls or boxes was set into it. The presence of a rectangular or oblong building dating to the early part of the Late Iron Age (based on preliminary pottery examination) is suggested as occupying much of the interior of the earlier roundhouse. Part of a very substantial flag floor associated with this building was uncovered. Finds

from these layers included pot-lids, a steatite spindle whorl and animal bone.

Part of the E sector of the massive roundhouse wall has been removed in antiquity and against the massive truncation front that resulted from this has been laid a revetment wall built in small, separate, non-integrated sections that hold back the mass of rubble. Floor deposits associated with this wall contained items such as a carefully made sandstone spindle whorl and an iron knife blade/small spearhead.

Structures E of the roundhouse/possible Iron Age settlement. A narrow extension trench, 25m in length, and 1.5m wide, on the eastern side of the main trench revealed that archaeological remains extended beyond the obviously artificial part of the mound and down the slope of the ‘natural’ rise at least as far as the eastern end of the trench, and these almost certainly extend beyond this, indicating a very substantial settlement. Geophysics indicates this could be as much as 70m in diameter. This trench contained a series of deposits and stone features stepping down the rise. These remains may be the upper elements or secondary phases of extramural buildings associated with a central roundhouse or broch. However, the presence of possible early Iron Age carinated pottery from these deposits and structures may indicate the presence of an early settlement pre-dating the roundhouse, or indeed that the origins of the roundhouse itself lie within the early part of the Iron Age. Towards the eastern end of the trench a midden area between two areas of walling was found to be very rich in animal bone, and this may be first indications of the upper fill of a surrounding ditch system.

Archive to be deposited with Orkney SMR and Orkney Museums.

Funder: Orkney Islands Council, Orkney Archaeological Trust, Glasgow University and Manchester University.

Excavation (7 August 2007 - 30 September 2007)

ND 4542 8688 Excavation continued this year from 7 August–30 September 2007 at the Cairns, Windwick Bay,

South Ronaldsay, as part of on-going research investigating Orcadian souterrains and their contexts. The main open area trench, some 15 x 15m, was the main focus of work.

Massive roundhouse. Work continued in structure A, a massive roundhouse with a total diameter of c22m, an internal diameter of 11.3m and a wall thickness of over 5m. A quadrant was excavated on the SW part of the interior, revealing that the wall of the structure survives to a height of 1.5m and that an intact flag floor survives beneath the mass of rubble which appears to have been deliberately and rapidly introduced at the end of the

life of the building. Occupational material and midden deposits in the interior yielded several coarse stone tools and pottery fragments of possible early Iron Age date. Several internal othostatic partitions indicate a radial layout inside the building. The excavated quadrant also allowed the examination of the previously recorded gap in the masonry wall of structure A and this gap was confirmed as a mural chamber, a teardrop-shaped cell around 1.8m in maximum diameter. This cell contained many large pieces of unworked whalebone. Also present was a large quantity of charcoal including possible charred withies together with a large piece of whalebone embedded in the wall at the entrance to the cell. This may indicate a hurdle screen or doorway that burnt down. Scorching on the adjacent inner wall face of the roundhouse may be related to such an event. It now therefore appears that the main entrance to Structure A lies further to the SE.

While the surviving height of the wall is too low at 1.5m to preserve evidence of some architectural features, such as a scarcement ledge, the question of the exact nature of this roundhouse was nevertheless further elucidated by an examination of the wall-core material on the NE where the wall had been slighted in antiquity. This revealed that the wall had an internal makeup of clay, soil and rubble dumps, which may indicate that the roundhouse does not possess the classic hollow-walled architecture of a ‘true broch’. This information,

together with the overall proportion of the wall thickness to internal area (the so-called percentage wall base), shows that structure A may exhibit some but not all of the traits of brochs. It is therefore likely to be a complex Atlantic roundhouse similar to Crosskirk in Caithness, rather than a high broch tower. The apparent early date of finds from the interior of the structure may support this provisional interpretation.

Sunken feature/Souterrain. Work continued in the centre of the roundhouse, in structure D, a rectangular sunken chamber that was partly revealed in 2006. A substantial deposit of in situ blocking stones was removed from the junction of the putative entrance passage and the chamber on the W side of structure D.

This revealed a large, partly rubble-filled, and still roofed, voided space defined by two lines of upright stone side panels, apparently confirming that this is a passage lying to the W of the main subterranean chamber. The fact that the structure represented a distinct, roofed building in its own right demonstrates that it was not simply a later modification of the interior of structure A, as previously considered possible. Structure D was constructed by digging into the mass of rubble in the interior of structure A and lining the cut with large uprights capped by substantial flags that have partly subsided, with a thick clay cap on top. In the deliberate blocking material we found a fragment of a copper alloy object, probably an unusual brooch. Below and in alignment with the base of the wall face of the souterrain chamber was a row of uprights which were probably further elements of the structure A internal architecture. This appears to show that the souterrain was built with some awareness of the layout of the massive roundhouse, and it may be that the deliberate in filling of the roundhouse and the construction of the souterrain were close in time, or indeed part of the same event.

Later activity on top of the remains of the roundhouse. Work also continued on the areas of later activity in the NW of the main trench. Structure B is a very well laid out set of uprights forming a rectangular or oblong building dating to the early part of the Late Iron Age. Its domestic nature appears to have been confirmed with the excavation of a large central rectangular hearth and the presence of occupation deposits. Stratigraphically, the construction of structure B was shown to be later than the use, and probably the abandonment, of the nearby souterrain. A small extension to the N of the main trench adjacent to structure B revealed part of another set of similar uprights laid out in identical fashion but turned around 90 degrees. This seems to represent the southern part of another building similar to structure B. A flagged passageway complete with threshold sill and pivot stone, possibly connecting these two structures as well as affording access to an external yard, was used at one point as a shell midden. This therefore appears to be a well preserved settlement zone of later middle Iron Age and/or earlier Late Iron Age date, post-dating structure A

extending across the centre and the N of the mound.

Archive to be deposited with Orkney SMR and Orkney Museums.

Funder: Orkney Islands Council, Orkney Archaeological Trust, Glasgow University, Manchester University.

Geophysical Survey (August 2009)

ND 4542 8688 An area of just under 1ha was surveyed in August 2009 using a fluxgate gradiometer around the

mound currently under excavation by Martin Carruthers. It was hoped that this would identify the extent of settlement activity on the N side of the site and detect any enclosure ditch.

The survey revealed that domestic activities extended c10m N of the large roundhouse or broch and that the

site was enclosed by a large ditch, most easily identified to the NW of the survey area. Extensive ridge and furrow was visible throughout the data and shares a scale and alignment with similar features already identified during excavation. Perhaps of most interest was a small mound c40–50m NE of the main site. The strength and form of the observed responses from the gradiometer survey clearly indicated that this had been an area of intensive domestic activity. It is unclear how this second area of activity related to the adjacent ‘broch’ site.

Archive: OCGU

Funder: Orkney College

Mary Saunders – Orkney College Geophysics Unit

Excavation (17 June 2009 - 12 July 2009)

ND 4542 8688 Excavation work continued 17 June–12 July 2009 as part of on-going research on the Orcadian Iron Age. The principal aims were to understand and further reveal a massive Atlantic roundhouse or broch-like building (Structure A) that makes up a large part of the mound and to continue to investigate the later Iron Age settlement (Structural areas B, C, D and E) that surmounts and intrudes into the building fabric of Structure A. The main focus of work was an open area trench of c20 x 20m.

The S and SW parts of Structure A, a massive roundhouse of over 22m in diameter, were further defined in an extensive band of work across the southern half of the building. In the SE the main entrance of the building was excavated.

During 2008, the upper deposits in the entrance had been found to consist of modern infill, and an episode of

antiquarian excavation was suspected. A complete excavation of this material confirmed this assessment. The excavation of the entrance area revealed a finely built passage surviving to a height of c1.5m and 1.3m wide. Slabs projecting from the wall faces form substantial door jambs, and a poorly preserved void on one face may indicate the presence of a bar-hole. The base of the passage was surfaced with a flag floor. The entrance is relatively simple in its construction with no side-cell or ‘guard chamber’ obvious as yet. A wall was uncovered at the inner end of the entrance passage that effectively blocked off the opening into Structure A. The stratigraphy indicates this wall was constructed at the same time that the interior of Structure A was deliberately infilled with rubble. At the southern (outer) end of the entrance passage another wall had also been constructed against the outer wall face of Structure A to block the entrance passage.

Whilst revealing extensive areas of the Structure A wall head, a well-built intramural gallery was revealed on the SW sector of the building. This T-shaped gallery ran S and N of a short entrance passage opening off the inner wall-face of Structure A. The northern part of the gallery contained the remains of a staircase. Only three stairs remained, the rest having been removed when large sections of the upper superstructure of the roundhouse were reduced at the end of its life. Nevertheless, the remnant staircase indicates that Structure A must originally have been a building of some height. Given that the staircase begins at a height of c1.5m above the inner floor of the roundhouse, it indicates that the staircase originally rose to an upper floor in Structure A. Structure A would have been an architecturally complex building, but despite this, its rubble and clay wall-core, as revealed in 2007, shows that it does not possess all of the traits of a conventional ‘broch tower’.

The second major area investigated this season was the later Iron Age settlement, which covers large parts of the remains of the Atlantic roundhouse and which was clearly reduced in height in preparation for the establishment of the post-roundhouse settlement. The character of this later settlement is such that it extends around the circumference of the Structure A superstructure. In many places individual buildings intrude into the physical fabric of Structure A in the manner of bites taken out of an apple. While this appears to be very unusual in Orcadian terms, a similar arrangement of later Iron Age settlement in relation to brochs and other

substantial roundhouses is attested to in Caithness. At least three major structural groups radiate around the Atlantic roundhouse ruins. On the N and NW there is a substantial complex of buildings and features grouped as Structure B area, which lie across the rubble infill and over the wall head of Structure A, while in the NE Structure C is a large building that intrudes into the wall of the Atlantic roundhouse.

Structure E in the SW is again embedded in the remains of A. Excavation of the intramural staircase within Structure A, The Cairns Excavation of the Structure B area was extended N and W to allow further investigation of this complicated suite of remains. A myriad of new walls and cellular features were identified, as well as hearths, wall-piers and orthostatic partitions. Essentially, a large rectangular building or two such buildings seem to have been succeeded by a less regular series of structures with curvilinear architecture and further

cellular arrangements. Of note was the presence of a large pit that had been dug into the northern floor space of Structure B late in its life. The pit contained decorated spindlewhorls, antler mounts and an extensive assemblage of animal bone, including many cattle mandibles and red deer antler. The pit was sealed with a large rotary quern. Most remarkable of all, the pit contained a small anthropomorphic head with clearly carved eyes, nose and mouth.

Work continued in Structure C, a large sub-circular building, which was further defined in outline and shown to be complex and substantial. Of particular interest was the excavation of a sub-circular furnace or kiln feature with a flue arrangement, an associated clay floor and heat-affected deposits spread across the building. Finds included a clay mould possibly for a ring-headed pin, a tuyère fragment, copper alloy pins, crucible fragments and various ferrous objects of indeterminate nature.

Structure E was fully excavated. This large rectangular building, with a rounded gable end on the N, appears to

have been open on the southern end. No discernible floor material was observed and the building was partly taken apart to allow investigation of the section of Structure A that it was built into. This work led to the discovery of another building directly beneath it, which was laid out on the same alignment. The floor deposits of this building appear to be well preserved and excavation indicated a part earthen, part slabbed floor. A substantial central rectangular hearth was also uncovered. Exploration of this building will commence next season.

From the work on the later settlement, it can now be stated that there are probably at least six later buildings nestled into the bulk of the ruins of Structure A. These remains almost certainly constitute the uppermost suite of buildings to have been laid out on the mound at The Cairns. It is probable that an even more extensive set of earlier buildings lies beneath the later Iron Age settlement. Previous work established that archaeological remains extend at least 20m beyond the outer wall face of the Structure A Atlantic Roundhouse.

A geophysical survey was undertaken by OCGU beyond the main excavation trench to establish the extent of the archaeological remains and to ascertain whether an enclosure ditch might surround the Iron Age settlement.

Archive: Orkney SMR and Orkney Museums Service (intended)

Funder: Orkney Islands Council, Scapa Flow Landscape Partnership, and Cardiff University Archaeology Department

Martin Carruthers – Orkney College

Excavation (16 June 2010 - 11 July 2010)

ND 4542 8688 Excavation work continued 16 June–11 July

2010 as part of ongoing research on the Orcadian Iron Age.

The principal aims were to understand and further reveal a

massive Atlantic roundhouse or broch-like building (Structure

A) that makes up a large part of the mound and to continue

to investigate the later Iron Age settlement (Structural areas

B, C, D and E) that surmounts and intrudes into the building

fabric of Structure A. The main focus of work was an open

area trench of c25 x 25m and Structures C and E.

Structure A, a monumental thick-walled roundhouse 22m

in diameter, was further defined in an extensive band of work

across the SE portion of the building. Approximately a third of

the rubble fill in the S and E sides of the interior of Structure

A was excavated. This facilitated the concurrent excavation

of the SE-facing entrance of the building. During 2008 and

2009, the upper deposits in the entrance had been found

to constitute modern infill, and an episode of antiquarian

excavation was confirmed. This season, the excavation of the

entrance area revealed a finely built passage surviving to a

height of c1.5m and 1.3m wide with no ‘guard-cell’ present.

The excavation of several tons of rubble from the SE portion

of the interior of Structure A revealed a series of orthostatic

stone fixtures, some of which were radial to the inner wallface

while others defined irregular-shaped spaces towards

the centre of the internal space. There were some indications

of later modification of the fixtures, indicative of at least

some complexity in phasing. This chronological complexity

was further supported by the presence of stone-built features

located just within the entrance that would have blocked

routine ease-of-access and therefore may well relate to later

use of the building. The revealed wall-faces showed signs of

several discrete areas of scorching and areas of adjacent ash

and charcoal spreads, again indicative of interesting events

late in the life of the building that will be further explored

next season.

View inside souterrain passage, Structure F, The Cairns, Windwick Bay

Excavation at the outer end of the Structure A entrance

revealed a finely built and still roofed stone-lined souterrain

passage (Structure F) that had clearly been built after the

main period of use of Structure A. This c3m long curving

souterrain passage connects up with the outer end of the

Atlantic roundhouse entrance passage, reusing it as the

chamber for a souterrain. This remarkable reuse explains

why the inner end of the Structure A entrance passage had

been blocked off by a substantial wall during the episode

of infilling and abandonment of the interior of Structure

A. Effectively, this wall had formed the back (inner) wall

of the souterrain chamber. The souterrain passage was

equipped with a pair of small door-checks just in front of the

area where it converged with the former entrance passage,

apparently confirming that this was a passage intended as

a crawl space for humans, rather than a substantial drain.

A small portion of the fill of the Structure F passage was

excavated and a whale tooth pommel, possibly from a sword

handle, was found.

On the eastern side of the wall-head of Structure A, an

intramural cell of tear-drop shape in plan, was uncovered

partly overlain by the Late Iron Age Structure C building,

and opening off of the inner wall-face of the Atlantic

roundhouse. Within Structure A this takes the total of such

cells to three. Taken together with the intramural staircase,

that was uncovered in the previous season these features

indicate the architectural complexity of Structure A. This

adds weight to the probability of it being part of the complex

Atlantic Roundhouse tradition, rather than the simpler form

of Orcadian substantial roundhouse.

Work also continued on the later post-Atlantic roundhouse

suite of buildings and activity areas. Work was primarily

carried out in Structures C, ‘the workshop’ area, and E,

another later Iron Age cellular, probably domestic, building.

Structure C, a fairly large sub-oval building, had previously

been found to have been constructed into the ruins of the

Atlantic roundhouse on its NE side. Its identification as a

workshop was further strengthened as it yielded several

heat-affected surfaces and features probably relating to

metalworking, as well as a fairly substantial assemblage

of items relating to metalwork production. These included

whetstones, mould fragments, copper alloy fragments and

large copper alloy sheeting. Of particular note was the

excavation of a substantial fragment of a small copper alloy

vessel, which, on typological grounds, has the appearance

of a plain hemispherical hanging-bowl from a set of balance

scales. To the author’s knowledge there are no other Late Iron

Age examples of such a device and it may in fact represent

a Scandinavian example. This, taken together with the

unusual incidence of organically tempered pottery (though

itself not unknown in Pictish assemblages) and steatite

vessel fragments from this building, may indicate a period of

early cultural interface between ‘Pictish’ and Scandinavian

groups focused around this particular structure. Provisional

archaeomagnetic dates for certain heat-affected features

late in the Structure C sequence indicate last heating of not

later than about AD 600 and therefore any such cultural

interaction is indeed of a precociously early date. Other

elements explored in this building included a well preserved

corn-dryer structure and the unusual survival of what may

have been either a peat stack for fuel or a turf wall.

Work also continued in Structure E, a multi-cellular

structure on the southern side of the main trench, which

had again been constructed by building into the remains

of the wall fabric of the ruinous Structure A roundhouse.

Excavation of a centrally-placed and finely built rectangular

hearth afforded the opportunity to subject the hearth deposits

to archaeomagnetic dating, which supported the Late Iron

Age attribution for this building, returning a provisional date

for the last time of heating of between AD 645 and AD 795.

The building also yielded small finds of polished pebbles as

possible gaming counters, and a fine copper alloy sewing


Finally, in the previous season a magnetometer survey had

indicated a separate geophysical anomaly lying to the NE

of the main Iron Age settlement. This c30m diameter subcircular

feature was sampled by trial trenching. Two small

trenches revealed a yellow clay surface or possible cap,

one of which contained two spreads of pottery of probable

Neolithic date.

Archive: Orkney Museums Service and Orkney SMR (intended)

Funder: Orkney Islands Council, Scapa Flow Landscape Partnership,

University of Glasgow and University of Aberdeen.

Excavation (18 June 2012 - 13 July 2012)

ND 4542 8688 Excavation work continued 18 June – 13 July 2012 as part of ongoing research on the Orcadian Iron Age. The main open area trench, c25 x 25m, was the focus of work with small extensions on the SE and W sides. The principal aims were to extend understanding of a massive Atlantic roundhouse or broch-like building (Structure A) that makes up a large part of the mound, and to continue to investigate the later Iron Age settlement (Structural areas B, C, D and E) that surmounts and intrudes into the building fabric of Structure A.

Structure A, a monumental thick-walled roundhouse, c22m in diameter, was further defined and explored both internally and externally. A trench extension on the W side was dug to characterise the outer wall face and the nature of deposits abutting it. This revealed that the broch walling was very substantial. It was composed of massive blocks in this sector, possibly a measure taken to strengthen the wall, since this section is adjacent to an intramural staircase. The wall had been placed directly onto the natural glacial clay as seen previously in a section against the N wall face of the broch and indicates that whatever old ground surface, topsoil and deposits lay there were scalped away in preparation of the broch foundation. Significantly, the level at which the broch foundations were present and at which the surface of the natural occurred in the W section was within a few centimetres of the same in the N section. This was despite the differential heights at which the two sections were made relative to each other and to the steep hillslope on which the site sits. This means that substantial earthmoving (terracing) was probably undertaken to provide a level platform for the broch on the break of slope.

Approximately two thirds of the rubble fill of the interior of Structure A was excavated, on its E side. This revealed extensive well preserved occupation deposits across the excavated area and substantial in situ stone orthostatic fixtures and fittings demarcating the internal layout of the building, at least in its latter stages of occupation. The occupation material will be excavated in 2013. The preliminary indications are that there is an extensive artefact rich and deep set of deposits representing a potentially long period of use in the building. The uppermost surface of the deposits is 0.5m above the floor level reached in the adjacent broch entrance, which itself is not necessarily the earliest floor.

The ongoing investigations into post broch activity on the site work focused on defining the stratigraphic relationship between the souterrains (Structure F) and a nearby Late Iron Age building (Structure E). The souterrain, a c3m long curving passage that reused the remains of the entrance passage of the broch, was found to predate Structure E and relates to a contemporary above ground building, thought to be a roundhouse, Structure G. The exploration of a quarter of this putative roundhouse revealed a well constructed hearth and some internal divisions or kerbing. The souterrain would have been entered from the N part of the above ground building close to its inner wall face, in a position and manner seen in other Orcadian souterrains/earthhouses.

The possible roundhouse was found to have been deliberately infilled with midden, presumably derived from Structure E immediately to the NW, and this infill also seems to have sealed the entrance to the souterrain at the same time. This midden was dated to the 7th century AD by the presence of characteristic ‘hipped pins’, and this fits well with the previously obtained archaeomagnetic date for a hearth in Structure E that indicated it was last heated between AD645 and AD795.

A clay cap was excavated from over the surface of the stone roof lintels of the souterrain and a series of rotary quern fragments were found to have been installed in a specially made stone setting or kerb, laid out on the roof. It may be significant that the central hole of the uppermost rotary quern was positioned over a gap between two of the souterrain roof lintels, encouraging the idea that it was intentionally set to allow some form of communication, either of the verbal variety or of materials, such as libations, from above to below. The full excavation of the floor deposits inside the souterrain is scheduled for 2013.

Work also continued on the later post-broch suite of buildings and activity areas in the N area of the main trench. Attention was focused on Structure C ‘the workshop’, and ‘Room 1’ of Structure B, a complex multi-phase arrangement of rectilinear and cellular wall lines, features and adjoining rooms or, more probably, entire buildings.

Structure C, a relatively large sub-circular building, had previously been found to have been constructed into the ruins of Structure A on its NE side. The identification of Structure C as a workshop was further strengthened by the presence of more heat-affected surfaces and features probably relating to metalworking, as well as a fairly substantial assemblage of items relating to metalwork production, including moulds of clay and stone and specimens of bog ore. More steatite finds, a vessel fragment and a spindle-whorl, may support the possibility that there is Late Iron Age/Viking interface present in this building. On the E edge of the building, a newly uncovered L-shaped portion of walling represents the remains of another structure, which post-dates the workshop and partially truncates it. Further excavation will be required to establish whether this building fragment is part of a Viking period structure.

Work in Structure B ‘Room 1’ revealed a large and elaborate long hearth c2.6m in length, which adjoins another hearth previously recorded in this room. The form and proportions of the hearth could indicate that it is an early Viking feature and this raises the possibility of a Late Iron Age/Viking interface within this building also. Samples taken from the hearth for archaeomagnetic and radiocarbon dating will hopefully elucidate this possible early Viking presence on site.

In summary, there is now confirmation of a Middle Iron Age period broch and associated extramural structures, followed by an immediately post-broch episode, involving the souterrain and putative contemporary roundhouse, itself succeeded by the extensive Late Iron Age settlement strung out across the remains of the broch. Finally, this may itself have been the subject of Viking interest possibly in the late 8th or early 9th century AD.

Archive: Orkney Museums Service and Orkney SMR (intended)

Funder: Orkney Islands Council, Orkney College UHI and ORCA

Martin Carruthers, Orkney College, UHI


Excavation (17 June 2013 - 12 July 2013)

ND 4542 8688 Excavation work continued, 17 June – 12 July 2013, as part of ongoing research on the Orcadian Iron Age. The main open area trench, c25 x 25m, was the focus of work with an extension on the S side. The principal aims were to extend understanding of a massive Atlantic roundhouse or broch-like building (Structure A) that makes up a large part of the mound, and to continue to investigate the area immediately above a well preserved souterrain (Structure F).

Structure A, a monumental thick-walled roundhouse, c22m in diameter, was further defined and explored. A trench extension on the S side was dug to characterise the outer wall face and the nature of deposits resting against it. This section was designed to be contiguous with one made immediately to the W in 2012. The 2013 trench recorded very substantial walling composed of massive blocks, very neatly constructed. Attempts to reach the base of this wall were abandoned when part of a possible new structure were encountered. This was well constructed using evenly coursed masonry, and the presence of substantial uprights, radial to its inner wall face seem to indicate internal partitions within a building. The new structure was itself sealed by material banked against the broch wall that is known to predate the establishment of the Late Iron Age settlement on top of the broch. While this structure therefore post-dates the construction of the broch in stratigraphic terms, it is a very good candidate for an extramural building associated with the period of the broch and is certainly the earliest building so far identified on site, after the broch itself.

Archive: Orkney HER and Orkney Museums Service (intended)

Martin Carruthers, Orkney College, UHI, 2013

(Source: DES)

Geophysical Survey (6 May 2014 - 8 May 2014)

ND 4542 8688 A 5.3ha fluxgate gradiometry survey centred on The Cairns was carried out, 6–8 May 2014. Excavations at this massive Atlantic roundhouse have been running since 2006, and this survey aimed to place the monument in a wider context. The results have highlighted the presence of numerous features of archaeological interest.

Although the NE limb (1) of a ditch surrounding The Cairns is clearly visible, no clear indication of an expected continuation to the SE side of the monument was found. Instead an area of strong and complex magnetic anomalies (2) indicative of enhanced soils related to domestic activity was

seen to spread 25m S to the front of the monument. Within this area, a peculiar linear feature (3) running SE–NW can be seen leading toward The Cairns.

On a rise 50m to the NE of The Cairns the survey shows a well defined area (4), 20 x 30m, where the very strong magnetic variations are indicative of domestic/settlement activities. Recent test pits excavated during student fieldwork conducted by UHI have discovered evidence of Neolithic occupation in this area.

Two localised clusters of strong magnetic anomalies (5) and (6) are located 100m and 125m to the SE of The Cairns. It is not possible to relate these possible features to the roundhouse.

More recent features are also visible; ridge and furrow can be seen spreading across the whole survey area and the remains of a former field boundary (7), 100m S of The Cairns, is seen running parallel to the actual field system.

Archive: RCAHMS (intended)

Funder: Orkney Island Council and Orkney Archaeological Society

Thomas Desalle – ORCA Geophysics

(Source: DES)

Excavation (15 June 2015 - 10 July 2015)

ND 4542 8688 Excavation work continued, 15 June – 10 July 2015, at The Cairns (known by RCAHMS as ‘The Cairns of the Bu’), as part of on-going research on the Orcadian Iron Age. The work focused on the main open area trench, c25 x 25m, and a 10 x 10m area in the northern extension immediately

to the NE of the main trench.

One aim was to enhance understanding of the massive Atlantic roundhouse or broch-like building (Structure A) that makes up a large part of the archaeological mound at The Cairns. Structure A, a monumental thick-walled roundhouse, possessing an overall diameter of c22m, was further defined

and explored internally. Substantial samples were taken from the deposits on the NE interior of the broch. A series of rough flagged surfaces were encountered similar to those seen in previous seasons further to the S. Beneath and to the W of the flags carbon rich soils and rough stone slab bases indicated occupation materials late in the life of the broch. Beneath these deposits a fairly sterile orange clay indicated a formally laid clay floor, and beneath this deposit lay more black organic-rich soil. Beneath this yet more sterile clay appeared to indicate an earlier laid floor surface, this time a saddle quern and quern rubber were found to have been deposited, placed on edge within the clay in a likely ‘structured deposit’. It appears that there is a succession of well-preserved formally laid-out floors and occupation deposits within the broch, at least in this northern sector, and this holds out great potential for assessing the uses of the broch from its beginning to the end.

The entire suite of deposits was sampled on a grid and will hopefully provide substantial volumes of environmental information and attest to the types of activities underway inside the broch through time. This information will be supplemented in future seasons when more of the interior of the broch will be excavated. One surprise discovery inside the broch was a very well-preserved built underground structure in the northern interior and very close to the inner

wall face. The structure is essentially of the ‘well’-type and has a set of some half a dozen stairs leading down to a partly rock-cut, partly finely coursed and corbelled chamber, c2m beneath the presently reached surface of the broch. Initial tests suggest that there is upwards of a metre depth of deposits in the base of this structure. The discovery of the ‘well’ is particularly welcome since one of the major aims of the project is to investigate the role and history of Iron Age subterranean structures and this feature will be fully excavated in future seasons.

In the northern extension trench (Trench M) work continued with the investigation of the large curvilinear feature that has been previously identified in geophysical survey and is thought to represent a large enclosure ditch surrounding the broch and settlement. The ditch was indeed located and, though c1.2m deep in this section, it was found to be very broad, and extended beyond the boundary of the extension trench. A network of drains was found to

sinuously cut through the boulder clay at the base of the ditch/gully. In several places where capping stones appear to have been lifted off in antiquity there were numerous edge-set quern rubbers placed in the sides of the drain gully forming its sidewalls.

In the western part of Trench M a series of building remains and features relating to activity after the ditch had been infilled were also investigated. The area was found to be as rich as the previous field season in evidence for metalworking. Around 30 fragments of well-preserved clay moulds were discovered and represent the debris from casting projecting ring-headed pins, rings and possibly penannular brooches. These were accompanied by metalwork itself including, broken copper alloy pins, and a ring. Importantly, crucible fragments and numerous pieces of slag and other residues and waste, as well as a large clay furnace feature itself, were found to suggest actual in situ metalworking was taking place in this corner of the site. The entire suite of materials is heavily indicative of a well-preserved and rare window on to bronze working at the end of the Middle Iron Age, which will continue to be a focus for excavation next season.

Excavation also continued to define the stratigraphic relationships between the souterrain, Structure F, and adjacent and overlying features. The southern, entrance end of the souterrain passage was found to extend further than previously thought and appears to make use of, or at least cut through, an older small circular building or cell composed of well-built orthostats.

Work also continued on the post-broch suite of buildings and activity areas in the western area of the main trench. Attention was primarily directed to Structure B, a complex multi-phase arrangement of rectilinear and cellular wall lines, features and adjoining rooms and discrete buildings. A well-preserved set of articulated red deer remains were found to partially underlie one of the hearths in this area and may represent a foundation deposit made during the initial stages of this phase of the use of the B area. Small finds from this area included a very finely carved conical gaming piece.

Archive: Orkney SMR and Orkney Museums Service (intended)

Funder: Orkney Islands Council and University of the Highlands and Islands Archaeology Institute

Martin Carruthers – Archaeology Institute, Orkney College, UHI

(Source: DES, Volume 16)

Excavation (13 June 2016 - 8 July 2016)

ND 4542 8688 (ND48NE 14) Excavation work continued, 13 June – 8 July 2016, at The Cairns (known by HES as ‘The Cairns of the Bu’), as part of on-going research on the Orcadian Iron Age. The work focused on the main area trench, c25 x 25m (Trench M), and a 10 x 10m area in the northern extension immediately to the NE of the main trench.

In 2016 a new zone of excavation was opened up between the existing trenches to create one extensive open-area. The new area was termed Trench Q and details of findings are given below.

Work continued on the inside of the broch (Structure A), and mainly concentrated on the SE quadrant largely within a well-defined area of the broch marked out by groups of orthostats that form partitions in the use of space. The removal of rubble on the N and W of the quadrant prior to

accessing occupation deposits revealed a finely built internal passageway, itself running off the main broch entrance. One side of this passage was constructed from large orthostatic panels while the other was a finely built coursed wall. This passageway had accessed two rooms set against the SE of the broch interior wall-face, and gaps in the orthostats defined

entrances into these spaces. The passage serves to further enhance the understanding of movement within the broch interior. The passageway and the rooms themselves formed the basis for the majority of the excavation work inside the broch this year. Within the passageway, occupation/floor

deposits were present across the surface and, in common with all sediments from inside the broch, these were excavated on a sample grid. Pottery and a fine copper alloy pin were among the small finds recovered.

In the SE zone of the broch interior a significant succession of deposits was observed ranging from dark organic carbon rich greasy layers to clean pale yellow clay. Significantly, the same detailed and quite clear stratification was present as had previously been witnessed in the floor/occupation deposits in the NE quadrant of the broch in 2015. It appears that similar, very good, preservation of floors is present across the whole of the E half of the broch thus far sampled.

Beyond the E zone of the broch, work continued in the western zone of the interior, and the last portion of rubble infill that had been deposited at the abandonment of the broch was finally excavated, revealing the full internal wall circumference of the broch upper levels. In the process, a doorway in the circuit leading into a very finely built, still fully roofed, extant intramural chamber, was discovered set in the western wall of the broch. Its entrance was similarly intact. This brings the total of substantial intramural features

in the broch wall to five. No excavation was undertaken within this chamber this season, but it has clearly been undisturbed since it was partly backfilled and sealed with rubble, sometime between the later 1st and 2nd centuries AD.

Work also took place on the outside of the broch, immediately against its outer wall and just to the N of the main broch entrance. Here, an area of rubble backfilled against the broch wall was sectioned to try to provide information with which to compare the dates of the backfilling episode of

the broch interior with that of the outside. Upon excavating the lower elements of this rubble, a remarkable deposit was encountered. A large, bucket-sized whalebone vessel carved from an entire vertebral bone had been deposited apparently immediately prior to the introduction of the rubble above.

Propped against the side of this vessel were two upright red deer antlers. The upper fill of the vessel contained a human mandible, and several loosely articulated animal bones (neonatal sheep). Finds of disarticulated or fragmentary human remains are a relatively common occurrence on

large Atlantic Iron Age sites and the concentration around the entrance area seems to have been a favoured zone for deposition of this kind of material, whether as foundation or decommissioning deposits. This particular, unusually complex, mixed deposit of artefacts and animal, and human remains, appears to represent something of an act of closure

marking the end of the broch immediately before it was shrouded in rubble externally. The human jaw, and a sample animal bone were radiocarbon dated and indicate that the person died around the same time that the deposition was made, sometime in the late 1st or 2nd century AD, very much in keeping with other C14 dates obtained for the end of the broch as an upstanding building. The dates also tend to support the idea that the broch was infilled internally and externally in the same single episode of wholesale decommissioning.

Immediately below the whalebone vessel deposit, a black, organic-rich deposit was encountered. This deposit was thick with limpet and winkle shells and replete with animal bone. This midden appears to emanate from the broch itself and the radiocarbon dates obtained from the deposits that sealed it suggest this is indeed material dating to immediately before the broch was abandoned, and possibly earlier.

It therefore appears that this midden will substantially assist the successful realisation of one of the key aims of the excavation of the broch: the detailed examination of a suite, or cycle, of production, processing, consumption and deposition, articulated across the broch interior, its floors, and its midden waste.

Additional work on the outside of the entrance area of the broch, also focused on the souterrain (Structure F), which post-dates the broch and which reused the broch entrance passage as the souterrain chamber. The fuller extent of the unroofed portion of the souterrain passage was tracked S, and found to form a c15m long structure. Meanwhile, work on the still-roofed portion, uncovered the entire run of surviving roof slabs and these were progressively recorded and lifted off. Three layers of major roof slabs were found to have been used to very soundly cap the souterrain passage.

The removal of these allowed a small amount of excavation inside the souterrain and on its deposits. The floors appeared to be simple and thin. A tiny amber bead fragment was recovered from one floor deposit during flotation of environmental samples. The floor/occupation deposits will be fully excavated next season. It was noted that, both the construction of the souterrain and the deposition of the whalebone vessel and human jaw appear to date from the same moment: immediately around the end of the broch, and may even have a shared significance in marking this seminal event. An antiquarian excavation (1901) in the entrance to the broch (later the souterrain chamber) found, amongst other things, two fragments of human bone.

In Trench M, the general aim was to gain more insight into the substantial metalworking that has previously been evidenced in this NE zone of the site. The remains of a clay furnace were fully excavated and found to be most likely involved in iron smelting. Slag and other residues indicated that there was heavy processing of iron ores under way. A second furnace was also identified and awaits full excavation.

As work progressed to reach earlier levels in Trench M, significant amounts of iron slag, bog ore, and so-called furnace bottoms, were encountered indicating that metalworking had been a hallmark of the N part of the site for a longer duration than the episode represented by deposits that have previously yielded considerable evidence of copper alloy

working including a spread of around 60 casting moulds that have been recovered in previous seasons.

Trench Q is the newly opened excavation area this year (c10 x 10m), positioned between the main broch trench and Trench M, and connecting both. Fairly substantial well-built wall-lines were identified in the western zone of this trench in particular. These were of double-faced character and in one case were surmounted by a smashed pottery vessel of a typical Middle Iron Age style. It seems likely that these walls represent elements of an extramural complex, or village, contemporary with the occupation of the broch.

Excavation work also continued on the so-called ‘Structure B area’, a complex of later Iron Age settlement post-dating the broch, and lying to the W of the broch, partly overlying the reduced broch wall-head, and formerly spilling over the infilled remains of the broch interior. The work identified a substantial entrance passageway leading into one of the B buildings, which had partly reused remnants of broch masonry to form on one sidewall of the passage. The base of the passageway constituted a set of shallow slab steps ascending in a southerly direction to give access to

Structure B1. The lowest base of the passage floor included four broken fragments of rotary quern which were found to refit together after they were recovered.

Archive: Orkney SMR and Orkney Museums Service (intended)

Funder: Orkney Islands Council and Archaeology Institute, University of the Highlands and Islands

Martin Carruthers - Archaeology Institute, Orkney College, UHI

Excavation (12 June 2017 - 7 July 2017)

ND 4542 8688 (ND48NE 14) Excavation work continued, 12 June – 7 July 2017, at The Cairns (known by HES as ‘The Cairns of Bu), as part of on-going research on the Orcadian Iron Age. The principal open-area trench, c25 x 25m, was one focus of the work, as was the NE extension area (Trenches M and Q), and a new extension on the SW side of the site.

Work continued on the broch (Structure A), and mainly concentrated on the interior of the western half of the broch. This was the first opportunity to properly explore this zone and to bring deposits here into phase with elsewhere in the broch interior. Beneath the uppermost rubble, initially, it seemed, the whole of the large western zone was one undifferentiated space with few substantial orthostats dividing it into smaller rooms as had been the case in the southern and eastern quadrants; however, it soon became apparent that there was further spatial complexity in the western zone and that the orthostats simply survived as a much lower preserved set of stumps than elsewhere.

Essentially the western area was divided, during later phases of occupation at any rate, into two rooms, one much larger than the other, with the hearth set centrally within the larger of the two. Nevertheless, the space available in the larger of the two rooms of the western zone was appreciably larger than any other of the other spaces within the interior of the broch previously encountered. The smaller of the two rooms also revealed the base of a gap and threshold sill and slab on its NE corner that communicated with the Northern room of the broch interior. This Northern room is the room that was previously found to contain the entrance into a ‘well’-type underground structure under the broch floors and deposits around the entrance to this were further investigated and found to be stratigraphically the same as the upper deposits in the western rooms. On the southern edge of the Western zone, close to the broch inner wall face and to the entrance of the still-roofed mural chamber discovered last season, the occupation deposits had been disturbed. A large pit c1.6m in diameter had been cut into the dark occupation deposit. The excavation of this feature is on-going and it may turn out to be a disturbed floor tank or similar feature.

Excavation of the actual ‘floor’ deposits of the western zone of the broch took on a now familiar character very much in accordance with those excavated elsewhere inside the broch. A very dark organic deposit was present across the whole of the Western zone and this was rich in well preserved environmental remains including faunal remains, fish bone and artefactual material such as pottery sherds and coarse stone tools. This deposit was the uppermost occupation material inside the broch, stratigraphically cognate with very similar deposits previously encountered to the E and S. The deposit should date to just before the end of the broch, probably around the mid-2nd century AD. Set within the matrix of this organic material was a fairly large,

heavily-cracked, slab that had obviously formed a late hearth setting. The hearth appeared to be sat upon a pronounced dome in the surface of the broch floor suggesting that it had been set up over a distinct mound of earlier deposits and quite possibly an earlier hearth, or hearths, and there were further hints of this situation in the form of edge-set stones and other heat-affected deposits emerging from beneath the edges of the hearth. The size and the form of this hearth are identical to another previously excavated in the SE room of the broch interior. Both hearths are from the same later phase and may encourage the idea that later in the life of the broch a possible concentric arrangement of space with a possible central focus had given way to a more binary division of the interior into two principal apartments. This changing spatial organisation would be familiar from other excavated brochs and roundhouses from the Northern Isles.

A final interesting aspect of the deposits in the Western zone of the broch again established familiar comparanda from the previously excavated zones. A small number of animal bone groups of loosely articulated mammal carcasses were encountered on the upper floor of the western zone. This chimes with elsewhere in the broch and appears to suggest that the final deposits across the greater part of the broch interior were of a formal, deliberate character related to decommissioning the broch.

This year further exploration was made in relation to the extramural settlement, or village, that lies around the outside of the broch. A substantial window onto this settlement was given by Trench Q to the N of the broch, as well as the SW extension. In Trench Q, a large volume of rubble and ashy soils effectively sealed and masked remains of village buildings here, and it took some time to reach the upper walls of these buildings.

Ultimately, the wall-tops of at least two, possibly three, Iron Age buildings were partially uncovered across Trench Q. One of these possessed a well-built, double wall face construction, and took the form of a semi-circular, curving wall and may turn out to be a roundhouse. This structure may well be part of a village complex contemporary with the broch itself. The partially revealed remains of another building were constructed by revetting into existing soils and rubble and may well be a building dating to later than the broch itself. Once again, there is a strong likelihood of a variety of building forms and styles present, reflecting the complexity and longevity of settlement on this side of the broch.

Excavation also continued outside the broch entrance on the E side of the main trench. Here also the well-built right angle corner of a building was identified. Based on the stratigraphy and the apparent orientation of the structure in relation to the broch, this may well be a portion of another building that is contemporary with the use of the broch.

A new trench extension (c8 x 12m) was made on the SW side of the site, in order to further contextualise the exterior of the broch and extramural structures previously observed on this side. Important information about the early stages of the establishment of the broch was gained. A large construction cut made into the natural glacial clay was encountered. This feature had been made in the form of a terrace made into the slope of the hillside, and while it had not been previously physically encountered, it had been hypothesised in order to account for differences in levels on site. The terrace cut also contained the walling of a building, which may well turn out to be part of the broch village on this southern side. Indeed, this situation may ultimately yield useful evidence for the contemporary relationship of broch village to the broch, and whether the extramural village was a planned entity with the broch from the outset of their construction.

The trench extension on the SW side of the site also brought to light part of the ditch system encircling the broch and village. This substantial cut feature was only partly excavated, but was found to be very rich in faunal remains and artefacts. Many sherds of typical broch period pottery with everted rims and globular bodies were recovered, as well as several bronze objects. One of which was a classic hand-pin and another a ‘long-pin’ with zoomorphic head.

One of the aims this season was to further explore the deposits inside the souterrain (Structure F) on the E side of the broch. Deposits that in-filled the souterrain were excavated and soil samples retrieved. Chemical investigation of some of these deposits may, it is hoped, shed light on any soil signatures left behind related to the peculiar aperture feature set up in the souterrain roof slabs whereby the setting of a pair of reused rotary querns provided a hole into the souterrain interior opening from the contemporary above ground surface. There are several instances of Iron Age querns reused in this way to form a porous lid on pits in wheelhouses in the Western Isles, and in one instance it has even been suggested that libations of some sort might have been poured through them.

In total, 17 Iron Age buildings have now been identified at The Cairns. Radiocarbon dating so far indicates a range in time from, at least, the 1st century BC to the 8th century AD or later. Several structures are now fully excavated and others will be sample excavated over the next two field seasons.

Archive: Orkney SMR and Orkney Museums Service (intended)

Funder: Orkney Islands Council and Archaeology Institute, UHI

Martin Carruthers – Archaeology Institute, Orkney College, UHI

(Source: DES, Volume 18)

Excavation (18 June 2018 - 13 July 2018)

ND 4542 8688 Excavation work continued, 18 June – 13 July 2018, at The Cairns (known by HES as ‘The Cairns of Bu’) as part of on-going research on the Orcadian Iron Age. The work focused on the principal open-area trench, c47 x 23m, Trenches Z, M and Q, and a further extension on the S side of the site.

Work continued on the broch (Structure A), across most of its interior, indeed, across a more continuous area of the interior than in any previous season. In the large western ‘room’ elements that were first revealed in the previous season, including a slab floor and a large hearth in the centre of the zone, were fully excavated and sampled. Beneath the slab floor and the hearth, lay yet another heavily heat-affected hearth slab on a slightly different alignment and a very dark, organic-rich occupation deposit surrounding it, filling much of the western room. Excavation of this organic occupation material, close to the western edge of the lower hearth, yielded a large complete annular glass bead of Guido Class 14 type, to which a 1st to 2nd-century AD date is appropriate. A globular fragment of Roman vessel glass was also found nearby, from the same occupation material. A large pit on the southern edge of the W room was excavated. It seems to have been created late in the life of the building, and was back-filled with rubble, earth and very little else. The excavated base and sides of the pit, however, appeared to reveal rubble from an early phase, beneath the settled floor surfaces, which are currently the subject of excavation. This suggests the possibility of lower deposits beneath a stony levelling event within the broch, and hints at an earlier sequence of occupation than the 1st and 2nd century AD exposed at this stage. Additional work in the W room involved the exploration of a large rubble filled void in the adjacent portion of the broch wall. This rapidly turned out to be a mural chamber, as suspected. This is the sixth, and probably final, mural chamber to be recognised. The chamber was 1.4 x 1.40m in diameter and sub-circular in plan. Excavation in the chamber proceeded until the uppermost occupation deposit, a red peat ash, was reached. It will be the subject of further investigation next season.

Excavation also continued in the NE room of the broch. The occupation deposits here, were composed of ashy material, and investigation revealed a complex sequence of floors and occupation, which were sampled for micromorphological analysis. The ashy soil chemistry appears to have preserved several fragments and splinters of unburnt wood. In the centre of the NE room a little pit, or posthole, was excavated, which contained small packing stones, edge-set around its edges. The ‘posthole’ had clearly been dug late in the occupation sequence, and had cut into earlier occupation deposits but it was in no way an early or primary feature of the broch layout. Indeed, the cut feature may be co-terminus with the previously excavated, very late, deposits related to fish processing in the NE room.

Work also continued in the S room of the broch, adjacent to the entrance to the mural staircase, and in the adjacent central passageway with which it communicates. Deposits here were notably dark and rich in carbonised organics. An array of bone objects, including a carefully trimmed and perforated antler object, stone tools, including a large inverted saddle quern, a range of ceramics, and a stone spindle whorl were recovered.

This season saw the full excavation of deposits from the ‘well’-type structure beneath the floor of the N room of the broch. Prior to discovery, during the 2015 season, the well entrance had been capped by a thin stone slab presumably shortly before the broch was abandoned. The slab was surmounted by just two thin lenses of broch occupation/activity indicating that it had remained accessible for much of the duration of the broch occupation. At some point, probably in the mid- to late 2nd century AD (as suggested by radiocarbon dating) the broch tower was decommissioned, reduced to about 2m in height, and much of the resulting rubble was used to pack the interior of the broch sealing it, and the well, under a thick blanket of stone.

Beneath the thin capping slab over the opening to the well, a quadrilateral entrance arrangement was formed by three side-slabs surmounted by coursed masonry creating a very well-constructed formal entrance, somewhat akin to a cist box. Below the level of this entrance landing, running from the open side (on the W), a set of seven finely-built stone stairs descend steeply with a pronounced leftwards (anti-clockwise) spiral to a rock cut chamber c1.06m wide. The stairs have been built against this rock-face. The entire length of the well, from entrance landing to the inner end of the chamber, is c1.6m. The well appears to have been constructed as a primary component of the construction of the broch and is substantially rock-cut (marks from metal chisels/chasers can clearly be seen in some of the faces of the rock). The builders then constructed masonry walls on top of the rock-cut ledges that were dug into the rock. Major bracing stones and lintels rise from these upper dry stone components to form a very well-constructed semi-corbelled roof. When first discovered, the well contained a substantial quantity of water, lapping at the uppermost stair.

In excavation, the uppermost deposits were very soft wet grey silt, and lay across the stone staircase and the interior of the chamber. The upper silts were largely sterile in terms of finds and any visible environmental remains. Whilst largely devoid of organics themselves, it is likely that these deposits acted as a seal and were instrumental in maintaining a thick seal on the basal deposits of the well, which turned out to be anaerobic in nature. The basal deposit of the well was a dark brown silt and dramatically different in colour and content from the overlying deposits, also giving off a distinctive aroma. The deposit was replete with preserved organics. Plant material including long thin heather roots and twigs, grass stems, mosses, and some insect remains were all visible and exceptionally well preserved. A thin piece of round wood with a carved head appears to have been a wooden peg (subsequently identified as willow). This object lay close to the base of the staircase and in a semi-upright position. It appeared to be in situ and pinning down some of the other organic materials. Further cleaning and sampling over the upper part of the deposit revealed the rim of a remarkably well preserved, very finely carved wooden bowl in the innermost, western, corner of the chamber. It was present in several substantial pieces, split horizontally in curving bands as well as in vertically fragmented pieces. The bowl is the subject of on-going conservation and assessment. It was made from half of an alder log and is c0.27m in diameter across the rim. In antiquity, the bowl was repaired in four separate areas, extending its life, and suggesting that it was a valued and curated object.

Following the recovery of the bowl, excavation of the lower portion of the deposit continued. Amongst the thick organic deposits, perfectly preserved bark, heather stems and moss, were present. A startling discovery in this material was about 20 fine pliable hair fibres. Remarkably, the hair has now been confirmed as human in origin. Further excavation of the fills uncovered more pieces of brushwood and these were noticeably thicker and longer. Indeed, they formed something of a structured appearance, with some pieces placed transversely across other ‘bundles’. Possibly this represented a brushwood, lining or filter at the base of the well, trapping sediments. During excavation, 100% of the sediment contents of the well were recovered, of which, 90 litres came from the anaerobic deposit, which will likely yield remarkably clear environmental insights. Full excavation revealed the rock-cut base of the well chamber as a gently sloping sub-oval platform of smooth rock. Overall, the well structure is some 2m deep.

Beyond the broch itself, excavation continued in the northern area of the main trench. This large area of contiguous buildings and features represents the northern portion of a multi-phase extramural complex surrounding the broch. Immediately outside and to the N of the broch entrance, work in Structure O revealed a large part of this substantial rectangular building. It was well-built with free-standing, double-faced, rather than revetted, walls, and could well be an important building close to the front of the broch. Structure O was sealed and partly filled by rubble emanating from the demolition of the broch in the late 1st to mid-2nd centuries AD, and is therefore certainly a Middle Iron Age building, although whether it is a primary component of the extramural village will only be revealed with further work planned for next season.

Directly to the NE of Structure O work also recommenced on another large extramural building, Structure Q. This building, the scene of work in previous seasons, is another elongated one, which excavation revealed to extend some 15m in length. At its southern end, Structure Q abuts the northern wall face of Structure O, but remains more vague and disrupted at its N end, due, in large part, to the thinner, less well-preserved stratigraphy at this edge of the site. Structure Q was largely filled with rubble and soil, which contained pottery, stone tools, and a small annular yellow glass bead of Guido’s Class 8 type, a type of bead with along currency. A small copper alloy spiral finger/toe ring was also found in this in-fill material. On its western flank the outer wall face of Structure Q is essentially the inner wall face of Structure K (the scene of metalworking), which it directly backs on to. An entrance connecting the two buildings was discovered on the W wall of Q. The entrance, flanked by a wall pier, was furnished with a well-built threshold with flagged floor surface and an upright sill. A large whalebone vertebra, which had initially been used as a chopping-board and then inverted was then utilised as a stout door pivot within the threshold area.

Work also recommenced inside Structure K, concentrating on the southern end of this building, which had been the scene of copper alloy metalworking late in its life. In excavating the fill of the southern end of the building two large re-fitting pieces of antler mount, with peg/rivet holes, were discovered that had clearly formed a handle plate of a large iron tool, and probably something like a substantial hand scythe.

Work on the southern extramural area of the site also continued this season. This concentrated on the previously identified ditch, which surrounds the settlement and one of the extramural buildings (Structure J), which sits snug against the southern exterior wall face of the broch. The main trench was extended on the S in order to provide a more substantial view of the ditch; however, the full exposure of the very broad ditch (in excess of 6m in diameter), was prevented due to the adjacent modern field boundary, an agricultural ditch and fence-line. Nevertheless, the excavation of upper ditch fills brought to light numerous animal bone groups, including large cattle and red deer bones, and extensive semi- and loosely-articulated portions of carcase, indicative of primary deposition, perhaps related to feasting and deposition of the refuse, in a likely structured manner.

Inside the enclosure on this southern side, two previously excavated sections against the southern outer wall face of the broch were connected up through the excavation of previous control baulks. This provided a complete overview of Structure J, confirming it to be a small kidney-shaped building and part of the extramural complex. It was clear that the building extended and contracted at different parts of its History. A complex sequence of phasing was also evident in the relationship between Structure J and a substantial revetment wall, itself part of the outworks of the settlement, and which has at some point been utilised as the western inner wall face of Structure J. Investigation of the southern section of this revetment showed that artefactual material, including a saddle quern rubber, was incorporated into its make up at the time of construction.

In total, 19 Iron Age buildings have now been identified. Radiocarbon dating so far indicates a range in time from, at least, the 1st century BC to the 8th century AD, or later. Several structures are now fully excavated and others will be completed over the next field seasons.

Archive: Orkney HER and Orkney Museums Service (intended)

Funder: Orkney Islands Council and Archaeology Institute, University of the Highlands and Islands

Martin Carruthers – Archaeology Institute, Orkney College, UHI

(Source DES Volume 19)


MyCanmore Image Contributions

Contribute an Image

MyCanmore Text Contributions