Font Size

100% 150% 200%

Background Colour

Default Contrast
Close Reset


Date 2016

Event ID 1040140

Category Recording

Type Excavation


NH 756 608 (centred on) The Rosemarkie Caves Project has been investigating the archaeology of 19 caves on the SE side of the Black Isle, since 2006. The project is run by a team of voluntary professional and amateur archaeologists, and is linked to the North of Scotland Archaeology Society (NOSAS).

Between 2011 and 2015 a programme of survey and test pitting provided baseline data of the sites’ archaeological potential. The project team conducted test pitting in eight of the caves, four of which were located in the Learnie group (1B, 2B, 3B and 3C). Stratified samples of animal bone and charcoal from archaeological layers in the test pits established 7th – 9th-century AD occupation in the Learnie cave group, with 2nd – 4th-century AD and 11th – 12th-century AD dates also derived from material in Learnie 3C. Significant evidence for 19th – 20th-century traveller occupation was found across the caves, particularly in the Learnie group and Ivy Cave, with frequent evidence for leather shoe making or repair, and the recycling of other types of materials. Learnie 2B was highlighted as a particularly intriguing cave given >1m depth of archaeological material and the presence of two buried mortared and stone-built entrance walls.

In 2016, open area excavations within Learnie 2B aimed to clarify the function and dates of occupation of the cave, and its importance in interpreting early medieval and post-medieval culture in the area. The most recent activity in the cave produced substantial evidence for late 19th- to early 20th-century occupation, including cobbled floors, a hearth and midden deposits containing a wide range of artefacts. The main occupation area was defined by a poorly constructed, breached, boulder wall. However, a more substantial wall blocked the entrance of the cave and was constructed during the late medieval period. It had been designed to hold a door, as revealed by the door checks and 1.3m wide entrance, and the walls, now 1m high, had probably stood at least 1.5m in height. There was also some evidence to suggest the presence of wattle screens with applied clay, which would have enclosed the cave more effectively, making the interior a darker space.

The archaeological deposits relating to the construction of the wall were found to be relatively sterile and failed to produce any evidence regarding site function. But it is clear that such a monumental feature was built for a specific purpose. Certainly it was intended to close-off the internal space of the cave from the outside world. It is possible that it may have been used for storage, for shelter or for gatherings, such as religious worship. The wall had been built over the top of earlier midden deposits, which produced several sherds of medieval redware ceramics, iron fittings and concretions, and butchered animal bone. The earliest deposits excavated within the cave, relating to the early medieval horizon, produced diagnostic features and materials to enable site function to be determined. Thin occupation floors interspersed by layers of relatively sterile sand suggest intermittent use of the cave at this time. These layers of material produced animal bone and shellfish, and also contained good evidence for metalworking.

The area of the metalworking activity was defined in the interior of the cave by post and stake-holes that most likely formed a wooden screen, closing off the back of the cave. Other elements of the metalworking infrastructure included a pit that most likely housed the furnace; a circular cobbled area adjacent to this that may have provided a stance for the operation of the bellows; a small hearth kerbed on two sides using small beach cobbles; and a pit containing large lumps of charcoal – possibly a charcoal store. The sandy floor layer forming the metalworking horizon produced slag deposits, plano-convex hearth base fragments and vitrified residues including one large clay fragment forming a part of the tuyère – the hole through which the bellows injected air into the hearth or furnace. Processing of the sediment samples taken in this area produced hammer scale and preliminary analysis of this material by Gemma Cruickshanks at NMS, suggests that bloom refining and smithing was potentially taking place in the cave.

The evidence for small-scale metalworking prompted the extension of the main trench into the dark alcove to the N. This involved the removal of the post-medieval cobbled floor and a sequence of midden deposits. Remarkably, the trench extension produced no metalworking residues, suggesting that the wooden screen (represented by the post and stakeholes) contained this spread of material. The amount of butchered animal bone increased dramatically in this area on reaching the early medieval horizon, some of which appeared to be cattle and formed distinct groups. However, cleaning back this material further revealed something completely unexpected: well-preserved human remains.

Excavation in the alcove revealed the outline of a large beach cobble, along with two articulated upper and lower leg bones. Other stones also started to appear in the area overlying the inhumation and revealed that these, like the larger beach cobble, had probably been used to pin down the body. Analysis of the distinct caches of butchered animal bone located above the cranium of the inhumation showed that these represented the main meat-bearing bones from eight adult cattle, while two bones from a horse were also included in the deposits. The bone groups may relate to a feasting deposit placed with the burial.

It was clear from the stratigraphic relationships that the individual dated at least to the early medieval period, but there was no diagnostic evidence such as grave goods to provide any further clues. However, evidence relating to the individual’s demise soon presented itself after revealing the skull. The roughly circular exit wound at the right temple area suggested that this was probably no accident! While informing the police and procurator fiscal regarding our discoveries, so that we could proceed with the recovery of the individual, we were also contacted by Sue Black, of the Centre for Anatomy and Human Identification, University of Dundee. Due to details revealed by some of our images, which had been submitted to her department by the procurator fiscal, Sue and her team kindly offered a full forensic analysis of the remains, along with a digital reconstruction of the individual’s face.

The individual was male, between 25–35 years of age, was Caucasian and stood between 5ft 6" and 5ft 9" high. Although some slight pathology was identified on his vertebrae to suggest possible mild osteoarthritis and some back pain, this was a robust and healthy individual in the prime of his life.

He had been well nourished and his teeth were excellent with no caries. He had very strong muscle attachments on his lower arms, capable of producing a strong grasp, which may be indicative of his work. There was no sign of systemic pathology, previous healed trauma, or disease. The only trauma that he had sustained was to his head. The first blow was to the right side of his mouth and it fractured teeth in the upper and lower mandible and displaced one tooth back into the bone due to the force inflicted. A tooth found during excavation in the thorax is a clear indication that he was still alive at this time as it is likely that he either inhaled or swallowed the fractured incisor. The second blow came to the left side of his jaw with some force. It severely fractured the left side of his chin and caused fractures on both sides of the jaw focused on the condylar process. The force of the blow also set off a radial fracture internally along the base of the skull. Next was a contact injury – probably caused by the individual falling backwards and striking his head on a hard surface – most likely as a result of impact numbers 1 and 2, which probably followed swiftly. While the individual was lying on the ground, a rounded implement was then pole-driven through the side of his head from left to right, just in front of the temple region. This set off fracturing around the face and skull and was most likely the cause of death. The final blow to the top of his head used tremendous force, creating a large penetrative wound which set up massive fracturing of the skull. This final blow would have been carried out while the man was prone and suggests significant ‘overkill’.

We have secured a total of six radiocarbon dates from the excavations at Learnie 2B and the date on the human rib overlaps with the dates from the main occupation/ metalworking horizon – although statistically the human remains appear to sit slightly earlier. Therefore, it appears that the body was placed in the cave immediately prior to, or contemporary with the start of the occupation/metalworking horizon. The date on the human rib bone provided a calibrated result of cal AD 430–631 at 2 sigma level of confidence (SUERC-70721).

Excavations conducted at the four caves in 2017 focused on finalising work in Learnie 2B and starting new open area trenches in Learnie 1A, 1B and 2C. Excavations in Learnie 2C were concluded and the results indicate that this long, dry fissure-type cave had been used only for post-medieval occupation. Excavations in Learnie 1A and 1B will be finalised in 2018, but the results so far indicate their use during the post-medieval, medieval and early medieval periods. Of particular interest are the rich occupation deposits revealed in Learnie 1B, dating to the early medieval period. This is the largest of the Rosemarkie Caves and we hope that evidence relating to site function will be revealed in 2018. The excavations bringing to a close our work in Learnie 2B included a trench investigating the relatively flat apron of ground just outside the cave entrance, and the removal of the final deposits in the back of the cave, adjacent to where we had recovered the human remains in 2016. Unfortunately, the excavations outside the cave failed to find any significant archaeological features or deposits. The deep sequence of deposits comprised residues redeposited outside the cave interspersed with large boulders (rock-fall from the cliff above the cave entrance) and wind-blown sand/sediments. The deposits did produce some metalworking residues, mixed midden material, and a number of artefacts of post-medieval date. However, the excavations in the back of the cave produced a final perplexing discovery.

Within the small excavation area, we uncovered two superimposed post-medieval hearths – the upper of which displayed the remains of a stone kerb on the NW arc. Removal of the ash deposits and kerb defined a rectangular stone setting, below which was revealed the almost complete articulated remains of a red deer… minus one half of the lower mandible and cranium. There were no obvious signs of butchery, while the stone setting did not contain the animal’s remains, but had been built over it. Removal of the rib cage revealed the complete leather sole of a shoe, while post medieval ceramics and glass sherds also accompanied the burial, proving its age in the overall sequence. The substantial quantities of leather shoe remains and leather off-cuts suggest this group of caves may have been used as temporary workshops, with a number of craft activities including basket-making potentially servicing this area of the Black Isle.

Archive: Rosemarkie Caves Project (currently) and NRHE (intended)

Funder: Highland Council, Cromarty Estate, CARD Fund, and Hunter Archaeological and Historical Trust

Steven Birch, Mary Peteranna and Simon Gunn – Rosemarkie Caves Project and NOSAS

(Source: DES, Volume 18)

People and Organisations