The Stonehall settlement – part three
By Sigurd Towrie
Excavated over a five-year period, the Stonehall Farm section was the most extensive area of the Neolithic settlement. 
That the large, low mound to the south-east of Stonehall Meadow was an area of occupation was hinted at by the ash, flint and stone tools and building stone unearthed during ploughing.
Excavation confirmed this to be the case, the evidence suggesting Stonehall Farm was “a third location of early Neolithic settlement running parallel with the occupation of Stonehall Knoll and Meadow.” 
The Stonehall Farm site was large, covering an area c50 metres east-to-west and 150m north-to-south, with the houses of Stonehall Knoll and Stonehall Meadow on its periphery. The entire occupation area was “a fully integrated settlement unit, creating a form of ‘neighbourhood’”, with communal activities taking place between the dwellings. 
Although radiocarbon dates showed all three Stonehall sites were contemporary, Stonehall Farm was markedly different to its neighbours. The mound consisted of a series of superimposed buildings – small and closer together – and surrounded by huge quantities of midden. 
This, it has been suggested, mirrored an Orkney-wide change in settlement patterns in the later Neolithic – a switch from the dispersed groups of dwellings to the close-knit, nucleated settlements typified by Skara Brae. 
Stonehall Farm allowed the excavators to chart this process of nucleation:
Up to this point, the end of the 4th millennium BC, the excavated evidence from Orkney suggests that settlement forms and architecture varied. But as we head into the 3rd millennium BC, there was “a convergence in the trajectories of virtually all Orcadian settlements”. When this “unification” occurs, the “inhabitation of Stonehall Farm changes radically in character.” 
Although only two buildings were fully excavated – Structure One and House One – there is little doubt that more lie to their west.
We will begin our exploration by looking at the scant remains of House Two.
This ruined building lay beneath Structure One and had been flattened to provide it with a stable base. Dating to around 3300BC, it survived as single wall section and a spread of collapsed masonry.
The pattern of collapse suggests House Two was rectangular and aligned south-east to north-west.
First encountered in 1995, Structure One was a D-shaped building incredibly similar to Structure Twenty-Six at the Ness of Brodgar – not only in their architectural features but the fact that both were poorly built.
Initially it was thought that One was a typical Late Neolithic house, but as excavation progressed that interpretation changed.
Although not certain, it is likely that Structure One began its life as a dwelling between 3300BC and 3000BC. At this stage, it had shoddily constructed entrance in the north-western wall and may have been circular in shape – if, as is thought, its straight south-western wall was inserted during later remodelling.
The building was erected on top of House Two and a partially levelled mound of midden. Stone orthostats not only defined its interior and furniture but, like Structure Twenty-Six, lined the inner wall face. We will return to these later.
A spread of flat stones at the building’s eastern end may represent the collapsed or dismantled remains of a “dresser” feature.
A paved cell in the south-eastern wall was particularly interesting.
It had no wall behind it, the rear formed by a stone slab propped against two blocks outside the building. This was hardly a sturdy structural feature, not to mention wind or watertight. So much so that the excavators pondered whether it was temporary, allowing access to the building from the south-east.
Its apparently shoddy construction, however, supported the evidence that Structure One was built on, and cut into, midden. The enveloping jacket of midden material would not only have insulated the cell but supported its flimsy rear wall.
A shallow stone box inserted into the floor adjacent to the northern section of the curved wall was found to contain flint and stone tools, including a large, polished-quartz axe, four well-used hammerstones and a Knap of Howar-type grinder.
After deposition this assemblage was sealed in the box by a large flagstone. This in turn was covered by a clay floor during the building’s remodelling.
The excavators considered the cache to be a foundation deposit relating to the rebuilding of Structure One. The poor bone preservation on site means that if it was accompanied by organic remains no trace remained.
Another interesting find was a pair of flint knives buried in occupation deposits that had accrued across the floor. The position of the blades suggests they had been buried in a bag or pouch.
Bringing in the dead
At some point in Structure One’s life its central hearth was replaced with a stone cist, its covering capstone resting on the floor.
This modification, the excavators declared, “drastically altered the character of Structure One.” 
If Structure One’s straight south-western wall was also added at this time it suggests the building had already fallen into ruin – or at least partially dismantled. Along the new wall – which was as shoddily built as the rest of the building – was a partially enclosed rectangular platform, shelving and a stone box. These, however, were in a poor state of repair.
Inside the cist were traces of poorly preserved bone, its condition such that it could not be identified. That said, it is more than likely the cist held human remains. 
In Skara Brae’s House Seven, a large cist beneath one of the “box beds” contained the skeletal remains of two females. Other human remains were found during the initial 19th century “clearing out” operations at the settlement.
In 1870, we find reference to:
While Laing makes no mention a cist, his account was based on a site visit and the viewing of Watt’s artefact collection. The fact the bone deposit lay beneath a paved floor is highly suggestive of a cist or, at the very least, a pit.
There, those seeking access to the south-western end of House Two had to pass over a covered pit beside the north-eastern hearth. This contained fragments of poorly preserved bone that wer too degraded to identify.
House Two was considered a site where the realms of the living and dead converged:
It was, suggested Professor Colin Richards:
More recently, a cist-like box above Structure Seventeen at the Ness of Brodgar was found to contain part of a human femur. The box, which had no capston, had been placed on the remains of Seventeen’s north-eastern wall section before being covered over and Structure Eight built on top.
Only the ball that connected the femur to the hipbone socket had survived. Whether more than just that fragment had been deposited remains unknown.
In all but one of the deposits mentioned contained disarticulated bone. Were skeletal parts being removed from chambered cairns, relocated and possibly exchanged? Perhaps to bring “life” to a new building, sealing agreements or even regarded as some form of “magical” protection?
While the cist deposits were buried or sealed, other accounts point at interaction with the dead being a part of daily Neolithic life.
At Skara Brae, Laing also referred to human bones in one of the house recesses:
One of these femurs, he noted, displayed cut marks, which Laing took to represent cannibalism.
If these “scores or notches” were indeed cut marks, it ties in with more recent work on Neolithic funerary practice. At the Quanterness chambered cairn the process of corpse decay may have been deliberately hastened. Examination of its bone assemblage showed some had cut and chop marks around areas of muscle attachment, implying the deliberate dismemberment of some corpses. 
Decomposition complete, evidence from across Orkney shows that bones in some cairns had clearly been reordered and rearranged. And, it seems, removed and relocated.
What was the reason for bringing the dead back into the world of the living? From ethnographic studies there were probably many – bringing life to a new building; sealing agreements or providing some form of “magical” protection.
But although we have examples of human remains in non-funerary contexts, the complete replacement of a hearth by a cist has only been encountered at Stonehall. 
At its most basic level the hearth represents the heart of a dwelling and its source of light and warmth. Its substitution with a burial cist, and its obvious connotations of death, must have been a powerful symbolic act.
Building a house in the Neolithic began with the laying of the hearth. Because they were re-used the hearth-stones probably came from a series of earlier dwellings. To the occupants of the new dwelling the fireplace was not only a tangible link to the past – to earlier structures and people – but also to future generations.
As symbols of continuity did the cist substitution represent the end of a line? Possibly, but were that the case we should surely expect to see more examples.
It is perhaps more likely that the introduction of the cist presaged a new role for the rebuilt and remodelled Structure One.
The cist was too small and too shallow to inter an entire corpse so was presumably built to house specific body parts. These may have had their own special significance, having multiple roles and various meanings, perhaps depending on the context and source of the skeletal material.
Given the lengths gone to contain the contents of chambered cairns, the incorporation of human remains into a structure or dwelling must surely have been controlled, with appropriate precautionary measures taken. These may have varied depending on the perceived potency of the deposit.
Within Structure One, the excavators pondered whether the cist was a means of safely bringing the dead into the domain of the living:
With this in mind it is interesting to note the stone uprights that lined the interior walls of Structure One. Reminiscent of the peristaliths  that surround the exteriors of some chambered cairns elsewhere in Britain and Ireland, Structure One’s uprights were suggested to represent “another mechanism of containing or wrapping the dead.” 
At the Ness of Brodgar, upright slabs also lined the walls of Structure Twenty-Six, but these were set at intervals around the perimeter and not continuous. If they were meant to contain something what it would have been is unknown. Twenty-Six’s hearth was intact and there was no cist.
Given the striking similarities between Structure One and Structure Twenty-Six, it is tempting to think they are roughly contemporary. Twenty-Six is late in the Ness of Brodgar sequence and thought to have been raised between 2800BC and 2500BC, using stone robbed from the initial demolition/rebuilding of Structure Ten and/or Structure Twelve.
This late date is problematic because birch charcoal from midden material in the Structure One cist produced a date of 3360-2900BC. In this case, however, the charcoal may relate to the removal of the hearth and the cutting into underlying midden to insert the cist. 
Although several contemporary buildings clustered around Structure One, only one was fully excavated.
House One lay to the north-east of Structure One and, like its neighbour, had been raised on earlier midden material. Its poorly preserved remains lay beneath rubble, soil and burnt material and the excavators suggested the building had been formally “closed” at the end of its life.
It had been a sub-circular building with internal features reminiscent of other Orcadian Neolithic dwellings.
Access was by a pair of entrances – one in its western side, the other in the north-eastern wall. A stone path led to the western main entrance, which, being offset, led to the right sight of the interior.
Immediately to the left on entry was a stone box feature that extended from the western wall almost to the hearth, thus blocking access to the left – northern – side. On the right side was an orthostatic box akin to those at Skara Brae, where they were interpreted as beds.
Its rectangular, stone hearth was roughly central and produced a date of 3350-2620BC. 
There may have been a stone “dresser” or box at the rear of the building, but the surviving evidence was inconclusive. Although not completely clear due to its ruinous state, House One appeared to have seen two phases of use.
-  Richards, C., Jones, R., Challands, A., Jeffrey, S., Jones, A. M., Jones, S. and Muir, T. (2016) At Stonehall Farm, Late Neolithic Life is Rubbish. In Richards, C. and Jones R. (eds) The Development of Neolithic House Societies in Orkney: Investigations in the Bay of Firth, Mainland, Orkney (1994–2014). Windgather Press.
-  Laing, S. (1870) On the age of the burgs or “broch” and some other prehistoric remains of Orkney and
- Caithness. Proceedings of the Society of Antiquaries of Scotland Vol 7.
-  Richards, C. (2005) The Ceremonial House Two. In Richards, C. (ed) Dwelling among the monuments: the Neolithic village of Barnhouse, Maeshowe passage grave and surrounding monuments at Stenness. McDonald Institute for Archaeological Research: Cambridge.
-  Crozier, R., Richards, C., Robertson, J. and Challands, A. (2016). Reorienting the Dead of Crossiecrown: Quanterness and Ramberry Head. In Richards C. and Jones, R (eds) The Development of Neolithic House Societies in Orkney. Windgather Press.
-  Where boulders or upright stones have been used to define the edge of a chambered cairn they are referred to as a kerb or peristalith. Sometimes the spaces between the large stones are infilled with drystone walling.
-  Richards, C. (2013) Wrapping the Hearth. In Richards, C. (ed) Building the Great Stone Circles of the North. Windgather Press, Oxford. Pp. 64–89.