Home About us Contact Us News Exhibitions Archive Links

Swartigill Excavation

De-turfed section of the north settlement

Eroding walling - 2004

During 2012 Yarrows Heritage Trust undertook phase 1 of an excavation of the Town ship of Swartigill. Phase 1 was to be limited to an initial evaluation of the archaeological potential of the site.

The name Swartigill comes from the old Norse (Black Pool Burn) and first appears in Thrumster Estate deeds from 1674 as the “Town and Lands of Swartigill” (From a disposition of the lands of Thrumster by John Sinclair of Brims to his son William).  However, no record exists of any leases throughout the 19th century, and presumably Swartigill, at some point, became part of the larger Raggra farm.


In the deed of sale to James Innes in 1819, there is no mention of it, but it reappears in a document in 1830, which may simply have been copied from an earlier source.

The Swartigill Burn is fed by sources which come together at the eastern end of the Oliclett Mesolithic site (ca 7-8,000 BC), an early hunting camp regularly visited by these early hunter-gatherer peoples.  The Swartigill burn was probably the route they used to reach these sites and flint debitage is found all along its length.

The site comprises vestiges of two settlements; one on the top of a plateau above the southern bank of the burn and one on the edge of the moorland to the north of the burn (HER record MHG 48275).

All that is visible are the remains of walls only a circa 0.3 m or so in height and largely covered in turf. Rig and furrow plough marks may be seen on all the flat grassland around the settlement.

A major boundary ditch and bank is still visible on the west side of the settlement and continues over the burn.  This could suggest a boundary between Raggra and Swartigill.

A third site comprising walling and rubble eroding out of the bank of the Swartigill Burn had been recorded by photography during the 2004 survey.  This site was and continues to be being, undercut by the burn and suffering erosion by sheltering sheep and consequently at risk. Large stones in the burn-bed suggest that much has already been lost.

The Burn site was re-visited in February 2007 for ground penetrating radar (GPR) survey of a small section of walling was drawn in section (HER record EHG 2647).  It was thought, at that time, that the burn site might be an old mill site; possibly a click mill. A preliminary and survey evaluation of the northern settlement and the rubble eroding from the burn was carried out 9th – 16th September 2012.

North settlement (HER record MHG 48275)

The settlement excavation indicate that building had stone built walls but there were indications that it may have been built on top of an earlier building possibly having turf built walls. A high proportion of bottle glass and a small makeshift hearth suggest that the site had been used as a picnic spot, probably by peat-cutters families during the 19th century. This observation is supported by local tradition. Further work is planned on this site to address some unusual features  and to confirm or otherwise evidence of an earlier structure.

Swartigill Burn site


This eroding site was first recorded by photography in 2004 (photos 1 to 5) walling was noted at both the eastern and western extremities of the eroding section which extended for circa 10 m along the southern bank of the Swartigill burn.


In 2007 CFA archaeology and Dr Susan Ovenden (Rose Geophysical Consultants) were conducting tests on ground penetrating radar (GPR) on wetland sites in the Yarrows area. The opportunity was taken to run the GPR over the burn site; Dr Mike Cressey of CFA also cleaned back and recorded a small section of the structure depicted in photo 5.


The objective of the evaluation of this site was to clean-back and record the full eroding section. It was initially thought that the remains represented a dam and / or mill. The box section feature is could be Iron Age but it is high in the section and could be a 19th century French drain (see Figure 3).

It was also noted that the rubble spread extended considerably further west along the southern bank, this exposure is currently stable and is quite distinct from the natural geology (photo 6). The potentially large area of rubble spread suggested that this might be the site of the original township of Swartigill.

The results of the cleaning- back were totally unexpected. Pottery recover from the erosion surface comprised fabrics consistent with the early Iron Age (c 0.72 Kgs). Finer quality pottery was also found that was consistent with a late Iron Age date; possibly Pictish (c 50 gms). Current knowledge holds that these pottery types have been exclusively associated with broch sites (pers. com Dr Andy Heald). None of the sherds were decorated; the author holds the view that the finger impressions under the rim of the Early Iron Age material results from the process of everting the rim and is not decoration.

The early Iron Age pottery was found close to the eastern extremity wall. This wall appears to be a substantial double skinned wall possibly having a rubble filled cavity, similar to the core of Thrumster Mains Broch excavated in 2012. The building stone also appeared to be similar to Thrumster broch and may have been sourced from the same quarry. The finer quality pottery was found close the western extremity structure; the two pottery types were not found to be intermixed (Figure 3).

It would appear that, to date, the ceramic assemblage from this site is known only from broch sites (pers. com. Dr Andy Heald).

Initial thoughts – the site in context

There would seem to be some similarities between the apparent double wall at the eastern extremity of the erosion face (photo 1) and the core of Thrumster Mains broch wall, and similarities of building stone at both sites. It seems likely that both buildings were built from stone quarried from the low ridge that runs under Thrumster Mains broch.

The positioning of the two sites are markedly different; the broch being located on a low ridge with commanding view-sheds while burn sit is inconspicuously located on the fertile first terrace of the Swartigill burn a relict melt-water channel.

The author is of the opinion that the origin of the feudal system recognised in the early medieval period may have its roots in the late Bronze Age / early Iron Age. I this proves to be the case then it is conceivable that the two buildings may represent a Caput and its associated Demesne; the core of a wider social and economic unit.

Alternatively the apparent double wall may be a misinterpretation of the

visible structure and may not be related to the ceramic assemblage.

Proposed next step

This site has the advantages that it lies relatively close to the surface and has not been disturbed by 19th century antiquarian excavation. It has almost certainly been robed out to some extent by the builders of the township of Swartigill; however it would appear that much of the Iron Age structure and deposits could remain undisturbed.

It is proposed that:

 A geophysical survey of the area is conducted to attempt to establish the extent and nature of the site.

 Test pitting be conducted verify the geophysical results.

 Evaluation of the results and assessment of the importance of the site to current research agendas, with a view to further work.

Funding be sought to finance the current proposals and any future work.

Plan of the cleaned-back eroding rubble

Back to Projects