We recently received 6 new C14 dating results and there is now a new basic chronology of activity in Learnie Caves 1A, 1B and 2B, revising some of the earlier phasing set out in the 2018 Data Structure Report. These three caves were the focus of our excavations in 2016, 2017 and 2018

In particular, we now have Roman Iron Age dates for Cave 1B (possibly relating to the very early Pictish period), pre-dating the placement of Rosemarkie Man in Cave 2B. This early activity in Cave 1B is contemporary with dating results from Cave 3C, Three Peaks Cave and Caird's Cave (with Caird's also yielding even older dates, as early as 300 BC).

The early activity in Cave 1B includes post settings (suggesting a wooden structure in the cave) and traces of iron metalworking (hammerscale and some vitrified furnace wall). The period is marked as Phase 1 in the table below. This period seems to have ended with a very hot burning event which burnt down internal wooden structure, scorched stones and paving, and left burnt halos of ash around the post settings. A period of abandonment followed.

Above: Phase 1 deposits in Cave 1B relating to the Iron Age.

Potentially contemporary with the burning event in 1B, or coming shortly afterwards, was the killing and placing of Rosemarkie Man in Cave 2B, along with the overlying feasting deposits marking his demise (Phase 2 dates in the table). This raises questions regarding any potential links with these two major events.

Phase	Date (AD)	Cave 2B	Cave 1B	Cave 1A	Period
1	133-329		Burnt bone		Roman Iron Age
1	257-414		Hazel charcoal		Roman Iron Age
1	284-440		Animal bone		Roman Iron Age
2	427-580	Human bone (Rosemarkie Man)			Late Iron Age/ Early Medieval
2	430-631	Human bone (Rosemarkie Man)			Late Iron Age/ Early Medieval
2	390-539	Cow radius fragment			Late Iron Age/ Early Medieval
2	415-550	Horse scapula fragment			Late Iron Age/ Early Medieval
3	11 dates between 636 & 941	charcoal & animal bone	charcoal & animal bone	charcoal & animal bone	Early Medieval
4	1443-1624		Semi-articulated poleaxed horse		Medieval/ Late Medieval
4	1495-1799		Sheep rib		Medieval/ Late Medieval
5	1658-1807	Cow rib from cist structure			Late Medieval/ Post Medieval
5	1681-1939	Bone fragment pre-dating construction of wall			Late Medieval/ Post Medieval

Above: The skeleton of Rosemarkie Man with associated deposits of animal bone in Cave 2B - Phase 2. Was this burial contemporary with the major burning event and abandonment in Cave 1B?

Activity then kicks off in all three caves at Learnie during the Early Medieval period with metalworking taking place in 1A and 2B and with traces of metalworking residues in 1B (Phase 3). The metalworking in all caves is followed by what appears to be occupation horizons, whether permanent or periodic. Re-occupation of Cave 1B sees much fire-cracked and scorched stone from burning event pushed towards the back of the occupation area. 

This occupation ends in all three caves during the later stages of the Early Medieval period. In Cave 2B this is identified by a fairly rich midden deposit with animal bone, while a more dramatic event appears to signal the closure of Cave 1B. Here the Early Medieval occupation horizons were sealed by a substantial stone deposit which contained the most abundant animal bone assemblage in any of the caves, generally comprising cattle and horse. We recorded similar closing-type deposits at High Pasture Cave (Skye) which may have marked significant feasting associated with these events.

Above: A section through the extensive stone deposit in Cave 1B, which seals the Early Medieval layers below (Phase 3). 

The medieval to late medieval/earliest post medieval horizon (Phase 4) is marked by the poleaxed horse remains in Cave 1B, which had been set down on top of the stone deposit forming the closing event described above. Several stone slabs covered the vertebral column of the horse remains, while other stones lay around the periphery forming a loose setting. In Cave 2B, a sheep rib bone recovered from a fairly rich bone midden was recovered from above the Early Medieval abandonment horizon. No dates for this period of activity was recovered from Cave 1A.

The earliest post medieval occupation deposits followed the above, with the large amounts of fish scales located in the deposits overlying the poleaxed horse; along with a wide range of artefacts. Interestingly, in Cave 2B, we found the cow burial within the capstone-covered cist structure (which lay directly underl a hearth setting). The date for the cow slightly overlaps with the poleaxed horse in 1B, but the horse was most likely deposited slightly earlier. All of the caves then display activity relating to use by Gypsy/Travellers.

Above: Poleaxed horse (Phase 4) and extensive fish scale deposits (Phase 5) in Cave 1B.

We have reviewed the artefact assemblage relating to these major phases of activity: Rich for the Post Medieval Gypsy/Traveller period; virtually nothing from the medieval/late medieval period with the exception of a few ceramic sherds from all three caves; then a few bone/stone tools, flint artefacts and pieces of pumice representing the Early Medieval activity.

In particular, there is very clear sequence in Cave 1B, with the burning event in the Late Iron Age, followed by Early Medieval/Pictish activity. Although on a much smaller scale, it has parallels with the sequence of activity uncovered during excavations at Craig Phadrig, Inverness - the destruction of the Iron Age fort (including vitrification), followed by the Early Medieval re-occupation of the hilltop. This is a sequence of events recorded at a number of other Iron Age/Pictish fort sites. The burning event in Cave 1B generated very high temperatures. The stone had not been vitrified, but much of the stone had been scorched to a red colour, obvious in the images from the excavations. There was also a lot of black and grey ash deposits associated with the burning. 

Above: Evidence of the significant burning event in Cave 1B.