Mill Road, Kingussie

NH 75679 00767 (centred) A desk-based assessment and watching brief was carried out at Mill Road, Kingussie, adjacent to St Columba’s cemetery. The work took place from December 2018 to October 2019. The watching brief recovered evidence for metalworking, two substantial walls most likely forming a part of the Carmelite Friary, an earlier fragment of wall possibly relating to an earlier church, and disarticulated human remains from within the walls of the church. Prior to the excavations associated with the watching brief in Mill Road, Kingussie, the exact location of the Carmelite Friary had only been identified through oral accounts.

The watching brief was conducted between 27 November and 6 December 2018. The pipeline trench was cut entirely through the tarmac road surface of Mill Road and comprised a narrow service trench 0.5m to 0.6m wide and up to 1.2m deep.

The first features to be identified were two drystone wall bases, set 5.2m apart. The SE wall was located below the tumble from an upper, later wall, and comprised a well-constructed drystone structure containing large to medium-sized rounded stone clasts (river cobbles), surviving up to two courses high and measuring 0.6m wide. The NW wall was also of drystone construction using river cobbles, surviving up to two courses high and measuring 0.7m wide. Both walls had been built on the natural ground surface. Metalworking residues recovered from between the walls in the service trench comprised just over 8kg of vitrified material deriving from different stages of the ironworking process; predominantly iron smelting with a small amount of bloom refining or blacksmithing evidence. A fragment of roasted ore and smelting slag indicate primary iron production was taking place, while hammerscale and slag spheres reveal blacksmithing. In addition to this, a fragment of unconsolidated iron and a key blank provide evidence of bloom refining and key-cutting in the vicinity. The abraded and fractured condition of the debris, along with the range of processes represented suggests the material comprises redeposited material within the structure. Two fragments of vitrified ceramic tuyère were also retrieved from the deposits, but these examples are too fragmentary and heat-distorted to be able to determine what form they were.

Other low rubble wall foundations were uncovered in the water main trench, but their function was difficult to interpret within the confined space. However, their morphology was very similar to those associated with the area of metalworking, so they may be contemporary in date.

Working to the NW, within the turning area at the end of Mill Road, the mechanical excavator bucket struck stone and in the next removal human bone was recovered. Excavation of the deposits below then proceeded using hand tools and after some initial cleaning back confirmed that the excavator had cut through the upper part of a large, mortared wall. From this point on to the termination of the pipe trench just beyond the turning circle, excavation revealed a second wall and other deposits of human bone.

The initial clean down of this section of the water main trench revealed a spread of buff to orange/brown sticky and firm clay containing a mixed, disarticulated human bone deposit, mainly comprising crania and long bones. This deposit was quite localised, running for approximately 1.2m along the trench to where it transitioned to darker silty sediments containing some stone. Further upstanding stone clasts were then encountered, some of which also displayed evidence of lime mortar bonding. Beyond these stones, the context transitioned again to mid-to-light brown/orange gritty sediments containing mainly small rounded stone clasts, but also with some larger, blocky clasts. This deposit was quite sterile and appeared to comprise water-derived material from flooding of the nearby Gynack Burn.

Working around the upstanding stone at the NW end of the trench, and removing the last of the overburden, two distinct faces of a large wall started to emerge. Aligned roughly N-S, the W face of the wall comprised some large angular and dressed stones interspersed with large river cobbles, while the E face also comprised some large blocks interspersed with smaller angular clasts. The distance between the wall faces measured up to 1.2m wide and the core of the wall comprised stone of varied size with patches of lime mortar bonding. Removal of the loose, gritty silts located to the W of wall face revealed up to three surviving courses of stone, including a basal course comprising large river cobbles, which protruded slightly outside the outer wall face.

The S wall of the church, which had been partially removed using the mechanical excavator, comprised several elements including slabs and blocks forming a N face with a buff, silty sediment matrix; a southern face comprising large and medium-sized river cobbles; and what appeared to be a secondary face or more structured core of the wall using river cobbles and more angular-shaped slabs, with a buff coloured, silty sediment core. Overall, the wall measured 1.2-1.3m wide between the inner and outer faces and stood up to four courses high (0.5-0.6m). At the limit of excavations below the S wall, stone slabs appear to have formed a foundation course. The two sections of wall revealed in the water main trench formed sections of the west gable and S wall of the church. The dimensions of the walls uncovered at Kingussie compare favourably with those investigated at the other Carmelite church sites in Scotland including Aberdeen, Linlithgow and Perth.

The upper, lime-mortared S wall of the friary appears to have been constructed over an earlier wall structure, the core of which is defined by the buff-coloured silty matrix – with no lime mortar present. Incorporated into the foundation courses of the S wall of the Carmelite Friary, it is possible that this wall fragment belongs to the earlier church that was said to stand on this site. During the excavation of this earlier wall, an alignment of three large basal stones was encountered with a gap to the west. This void was filled with dark brown silty sediments containing disarticulated human bones. The void most likely indicates the location of a doorway in the SW corner of the south church wall, or an entrance within the wall relating to the earlier church.

Working from the E face of the W gable wall in a SE direction down the pipe trench, excavation of the buff clay-like deposit revealed a complex of human remains including a significant number of crania and crania fragments, and long bones. The context also contained small to medium-sized stone clasts, which were mainly angular, but with some rounded river cobbles. At the SE limit of the deposit, the top of an upright slab was revealed measuring 0.4m long and 0.35m high which appeared to abut stones set on edge to the W and E, potentially forming a division wall. It is possible that the complex of human remains within this stone setting formed an ossuary/charnel-type deposit.

To the SE of the slab, the deposits comprised mid-to dark brown silty sediments containing angular and rounded stone clasts (mainly small), and with some charcoal flecks, fragments of lime mortar and human bone groups. The disarticulated human bone in this deposit was distributed at intervals along the trench and comprised crania and long bones, and displaying a distinct absence of other skeletal elements. The deposits within the walls of the church were removed down to the basal courses of the W and S walls, with disarticulated human bone including crania fragments recovered to these lowest levels.

Artefacts recovered from the deposits from within the church walls were few and mainly comprised iron nails, most of which had been bent (possibly during removal from timbers); and two very small sherds of medieval redware ceramics. The human bone was found to be in a very poor state of preservation, with many long bones fragmenting on lifting. No grave cuts were identified in either section of the water main trench where the human remains were recovered, while no remains of coffins or their associated furniture were found.

Analysis of the recovered human remains has identified a minimum number of 28 individuals of both sexes (including 8 males and 4 females) and varying ages (between 6 months and c.35 years). Very few pathological lesions were noted, in keeping with the poor condition of the remains. The major pathological traits included tooth loss and while the cause of tooth loss is not always clear, in most cases this was most likely caused by dental caries. Few teeth were recovered from the excavations, but showed that one individual had severe dental hypoplasia and showed repeated episodes of stress from the age of about 4 to 9 years of age. Dental hypoplasia can result from poor health or nutrition and this individual died either in adolescence or early adulthood. Three adults and one child had possible porotic hyperostosis, which can be attributed to anaemia during childhood. However, no cases of osteoarthritis were identified from any of the postcranial remains, which is probably reflective of the general younger adult age at death.

Two charcoal samples recovered from the metalworking residues were submitted for radiocarbon dating and returned calibrated dates of cal AD 775-965 and cal AD 891-1012. The presence of ironworking debris in contexts dating to between the late 8th and early 11th centuries AD can be paralleled at early medieval ecclesiastical centres elsewhere in Scotland, including workshops in Period 2 and 3 at Portmahomack and industrial areas in Periods I-IV at Whithorn. Large deposits of ironworking debris from this period have also been discovered during excavations on Iona and more recently a small quantity from the later ditch at Kinneddar, dating to the 9th to 11th centuries AD. The presence of ironworking at Kingussie from this period is a significant addition to this growing corpus, though the nature of its context and whether it is related to an earlier church remains unclear.

A total of 12 tooth and bone samples were selected from the human remains for radiocarbon dating and produced results ranging in date from cal AD 1439-1619 to cal AD 1493-1651; with many of the results relatively contemporary in date. The earlier range would fit in with the use of the Carmelite Friary, while the later date range may be associated with its abandonment. However, the disarticulated nature of the human remains including mainly crania fragments and long bones – and only a small number of post-cranial bones from other areas of the body – would suggest that these are potentially burials disturbed from other contexts and re-buried within the church. The nature of their re-internment within the confines of the church requires further analysis and interpretation, including comparanda from the other excavated Carmelite churches in Scotland, and elsewhere in the United Kingdom. However, it is possible that a considerable number of the remains relating to these individuals were deposited in the church at the time of, or after its abandonment.

Archive: Highland Council HER and NRHE (intended)

Funder: Scottish Water

Steven Birch - West Coast Archaeological Services

(Source: DES Vol 20)