In 2010, excavation work undertaken by Dr Clare Ellis of Argyll Archaeology in advance of the construction of Dunstaffnage Marine Science Park revealed a number of features, including a crouched infant inhumation burial, originally reported here. Initial assessment of the range of features present suggested that they were likely to be of Bronze Age date; however, radiocarbon dating has now been completed on a range of samples taken during the excavation, and these have produced unexpected results.
Although one pit produced a date of 1091-911 BC, falling within the expected late Bronze Age range, other activity from this period is limited to a couple of pits packed with fire-cracked pebbles. The restricted quantity of Late Bronze Age features is surprising given the presence of a significant number of features and structures of this period excavated in a field just to the south, and it is likely that the pit identified in this phase as part of a larger Bronze Age settlement and landscape.
Other samples indicated the presence of two separate periods of occupation, visible as two distinct clusters of features. Towards the northern end of the excavated area, an alignment of pits produced dates in the late Iron Age, predominantly falling within the 1st or 2nd centuries AD. This included the pit that contained the crouched inhumation burial, which produced a date of 96BC - 70AD. The function of this alignment remains unclear, though its east-west orientation may be significant, perhaps being intended to reflect the rise and setting of the sun. This may also explain the location of the infant burial at the eastern end of the alignment, a relative new born being associated with each new rising sun.
Samples taken from the southern end of the site showed evidence for occupation during the early medieval period, producing dates in the late 7th to late 9th century range. The site at Dunstaffnage, as excavated, comprised two granaries, a grain drying kiln, one or two pit basket boilers, two cobble hearths, a number of probable storage pits, a probable timber framed structure, a rubbish pit, probable fence lines and a few isolated postholes. In addition the isolated pit-hearths may have been located within timber and/or earthen structures that had little or no foundations and so all traces of these, bar the central pit-hearth, may have destroyed by subsequent ploughing. The main activity visible in the archaeological record is the preparation of oats and barley for long term storage. The dominance of oats in the cereal assemblage recovered from the fill of one of the corn-dryers would typically have suggested a Medieval or later date for the feature; however, samples taken from this returned a date in the range 671-869 AD, representing a particularly early occurance.
It is unclear whether that the Early Historic structures and deposits at Dunstaffnage are part of a permanent hamlet or represent the remains of seasonal agricultural activity. There is a distinct lack of the type of material in the form of stone tools, rotary querns, animal bone etc. that would normally imply more permanent settlement. It is generally assumed that during this period the upper levels of society would have occupied defended sites such as duns and forts, while some crannogs at least appear to have been seasonal specialised craft centres. The name 'Dunstaffnage' implies the presence of a dun, although no trace of it has been found. However, some 3.5 km to the south is the probable royal fort of Dunollie. This fort appears in the Iona Annals as Dun Ollaigh, being referred to in 686, 698, 701, 714 and 734; it was destroyed in 701 and rebuilt in 714 AD, and it is possible that the features at Dunstaffnage could represent the remains of a seasonally occupied farm under the control of Dunollie. Towards the end of this period, attacks on Iona by Norse invaders are documented in 795, 802, 806 and 825. Although there is no evidence for the deliberate conflagration of the Dunstaffnage site, a Norse presence is hinted at by the fact that a fragment of copper alloy Viking ring money was reportedly found by a metal dectorist in one corner of the excavation area prior to archaeological fieldwork taking place.