The cold, dark waters of Scottish lochs hold a certain place in the imagination – conjuring up images of the Loch Ness monster, or for others, the animistic water spirits of Scottish folklore known as ‘kelpies’, not to mention the stunning backdrop for numerous movies.
However, for a handful of of archaeologists, the dark water of Scottish lochs hold a different, more tangible riddle – island dwellings, typically represented by the crannog, a type of small artificial islet, and the Hebridean ‘island dun’.
While the concept of an island-dwelling is relatively straightforward, current terminology employs a number of overlapping labels. The Gaelic word crannog literally translates as ‘son of tree’ or ‘young tree’. However, the term also applies to numerous wooden tools or implements ranging from ship masts to butter churns no less, leaving us with open-ended meanings.
Though there are wide variations in methodology, the majority of mainland Scottish crannogs would have been constructed by driving timber piles into the loch bed, then filling in the interior with peat, brush, stones or timber until a solid foundation was formed. Today, most crannogs appear as little more than unassuming, heavily overgrown islets beyond the reach of grazing livestock, while in many instances, others exist as completely submerged stone-capped mounds, only accessible to intrepid, drysuit clad divers.
In largely treeless areas such as the Western Isles, these island dwellings take on a different character, employing a wide mix of natural, artificially enlarged or completely artificial islets. These unique Hebridean examples are traditionally referred to as ‘island duns’ due to their use of stone rather than timber – the same, identical concept as timber islets, simply made using what materials were available in the sparse Hebridean environment. Rather than timber roundhouses atop prehistoric crannogs, drystone architecture such as duns and brochs were typically constructed on natural islets in the Hebrides, as artificial foundations often proved incapable of supporting the immense weight of these robust stone monuments.
Largely free of the dense vegetation which characterises their mainland counterparts, Hebridean island dwellings tend to survive as prominent, easily recognisable stone remains of a broch, dun or later rectilinear Medieval ruin, usually connected to the shore by a slippery stone causeway which is sometimes slightly submerged – an approach often interpreted to make access difficult, though this submergence may simply reflect a change in loch level since construction.
Crannogs and island duns form a unique type of island dwelling which have a distinct distribution limited to Scotland and Ireland, along with a single, lone exception at Llangorse Lake, Wales. Continental comparisons exists, such as lake-side platforms found in Lac de Neuchâtel, Switzerland, though their free-standing design and overall chronology remains fundamentally distinct from their brethren across the North Sea.
During my PhD research on Scottish crannogs, I quickly realised that a single inventory of Scottish sites was non-existent. One of my first steps before heading to the field was to compile a gazetteer which eventually containing 577 of these island dwellings, the first to incorporate some 175 sites from the Western Isles into a comprehensive database alongside mainland crannogs . While this figure may seem large, this represents less than half of the Irish inventory, with no less than 1200 known examples for both the Republic of Ireland and Northern Ireland.
Surprisingly, despite a strong concentration of crannogs in south-west Scotland, no artificial islands have yet to be found in England, although sites at Glastonbury and the Somerset Meare appear to employ raised platforms in a wetland setting. While this overall distribution is perhaps only a reflection of modern political boundaries, it also highlights a cultural divide in the way our ancestors viewed their place in the world by living on a small islet, much in the same way the distribution of brochs is limited to Scotland. Nevertheless, it still remains perplexing that no crannogs exist in areas such as the English Lake District, where ideal conditions exist for their construction.
Radiocarbon determinations show that these enigmatic islets were constructed primarily during the Iron Age and Early Medieval Period, although exceptions abound. The overall limit of chronology ranges from the Neolithic site of Eilean Domhnuill or ‘Donald’s Island’ in the Western Isles of Scotland to Eadarloch, constructed during the 16th century AD near Fort William in the rugged Scottish Highlands, This heritage spans over five millennia, making crannogs one of the most enduring and unique site-types in European archaeology.
Crannogs are typically viewed by archaeologists as defensive retreats, lordly homes or monuments, based upon vague Medieval references and a small number of antiquarian excavations. However, the small number of modern crannog excavations largely reveal a more mundane existence as simple homesteads in later prehistory, with finds consisting largely of stone tools and wooden implements such as bowls, platters, utensils or ards – a rudimentary handheld plough. Pottery fragments – ubiquitous on most archaeological sites, are absent from mainland Scottish sites, occurring with any regularity only in the Western Isles.
One of the hallmarks of a crannog excavation is the exceptional level of organic preservation encountered due to the waterlogged conditions – wooden bowls, worked timbers, textiles, leather and ecofacts survive with virtually little or no deterioration. Instead of a faint stain in the soil marking the remains of timbers on ‘dry’ excavation sites, on crannogs, one frequently finds worked timbers entirely intact right down to the axe marks, while food-dishes and domestic containers may also contain preserved contents such as butter.
Despite their watery location which suggests strong defensive qualities, evidence of prehistoric violence, warfare or weaponry is almost non-existent. This contrasts sharply with frequent Medieval use as defended homesteads, often of a lordly, high-status nature. Indeed, this Medieval and Post-Medieval reputation as a refuge or battleground is much clearer than the prehistoric evidence suggests.
The unique illustrations from the Queen’s mapmaker, Richard Bartlett, who spent three turbulent years in Ireland from 1600-1603 surveying the Irish countryside for English troops, provides the only contemporary glimpse into this long-standing tradition. His depictions of English troops laying siege to a crannog are the only surviving images known. Ultimately, Bartlett went to his grave with his impression of these watery dwellings. Sir John Davies, who accompanied Bartlett, bluntly relates: “When he [Bartlett] came to Tyrconnell the inhabitants took off his head, because they would not have their country discovered” .
Though ‘living on a small artificial islet’ may sum up the overall mentality, beyond this general point it is difficult to apply a single label which can adequately describe the underlying rationale from the Neolithic Period to the Post-Medieval Period. Prehistoric examples, particularly from the Iron Age in Scotland (c.700BC to AD400), appear to have been laboriously constructed, then intermittently occupied for several centuries and abandoned, only to be re-occupied centuries later during the Early Historic and Medieval Periods. The initial impetus for crannog use during the Iron Age may be linked to Pagan beliefs which deified natural features in particular – pools, lakes and watery places. The use of an artificial islet may also hold a specific meaning, such as control over a territory, rather than simply occupying a nearby natural islet – a scenario which is often the case in large lochs such as Loch Lomond and Loch Awe in Scotland. Whatever the motive, it is clear that prehistoric occupants wanted to be on an artificial islet and not a natural one.
However, it appears that a second wave of crannog reoccupation and construction began sometime around the 6th Century AD, typified by sites such as Loch Glashan, Buiston and Ederline, Loch Awe. Mentions in the Irish Annals of battles at Ederline, and an increasingly number of radiocarbon determinations from this period broadly coincide with increasing large-scale unrest in south-western Scotland during the second half of the first millennium AD. It is difficult to glean much from these scant written references, but they suggest that crannogs were the focal point for important events, and were occupied by the upper levels of society, at at least during uncertain times.
Surviving Late-Medieval written references see crannogs and increasingly, small natural islets used as neutral ground for council or clan meetings and weddings, or as prisons or hideouts for ‘freebooters’ or thieves, or even a convenient place to keep hunting dogs. One example, Dun an Sticer in North Uist, initially held an Iron Age broch which was re-occupied in the 16th century for a short period by a would-be usurper who attempted to take control of the area. According to local lore, this unfortunate individual was betrayed by his mother as he swam away from the site, and was imprisoned only with salted beef and left to die of thirst . Therefore, use varied widely from site to site, while interpretations regarding use change just as frequently in the archaeological record, from humble domestic settings, to centres of violent conflict.
This reuse strongly suggests that the collective memory of these artificial islands remained very much alive in local tradition, while the identity of the former occupants remained a mystery – the stuff of tales and legends, however close they may be to the truth. Likewise, many crannogs give away their artificial nature away when loch levels drop during dry spells, revealing a complex series of sharpened timbers around the margins which consolidated the mass of material brought to construct the mound. For local inhabitants, the revelation that a near-by island was actually man-made would likely have been a surprising revelation, prompting much speculation and gossip regarding the history and former inhabitants of the islet.
All told, crannogs would have the seemingly ideal mix of mystery and intrigue to draw their attention to bards and story-tellers, who we can imagine spinning tales of ancient kings and sunken houses around the hearth to a wide-eyed audience. This mystic quality may have inspired local chieftains and lairds to reoccupy these sites long after their initial phase of use ended, perhaps in the hopes of projecting a sense of legitimacy on their own political designs. Other, more mundane factors may simply have been a convenient place to retreat when things became unsettled, thus reusing ancient places for more immediate purposes.
Today, crannogs remain relatively obscure outside of Scottish and Irish archaeology. Interpretations are reliant upon less than a dozen large-scale excavations since the 1850s, while many of these struggled to reach the earliest layers due to constant flooding which often becomes a battle between the on-site pump and the loch with predictable results. Only one crannog – Oakbank at Loch Tay, has been excavated extensively underwater , ultimately a time-consuming affair despite the calm conditions compared to offshore marine archaeology. However, bypassing the land/water restraints that plague exposed crannog excavations may yield superior results.
Meanwhile, many crannogs have not been properly surveyed or even visited by archaeologists due to their remote location, a factor which currently hinders future research designs. While my PhD research examined crannog and islet use over time in Scotland, I subsequently focused upon integrating sites in the Western Isles into the overall crannog record through a combination of desk-based research combined with investigations of some 50 sites. My initial dives in Hebridean lochs produced some exciting, largely-intact pottery finds which are currently being written up. I am planning extensive survey in the future in the Western Isles, to create a solid basis for future study. It is fair to say that despite 150 year of crannog research, these mysterious islets remain an enigma in regards to their prehistoric origins and meaning. Perhaps above all, it is refreshing to see that new avenues of archaeological enquiry still exist in an ever-shrinking world.
Written by Dr Robert Lenfert
Header Image : Island Dwellings 2: Dun Torcuill, North Uist. Most island duns remain in excellent overall condition due to their remote location. Credit : Dr Robert Lenfert
Lenfert, R. D. 2013. Integrating Crannogs and Hebridean Island Duns: Placing Scottish Island Dwellings
Into Context. Journal of Island and Coastal Archaeology. 18.1: 122-143. London: Taylor and Francis.
Andrew, J. H. 2008. The queen’s last map-maker: Richard Bartlett in Ireland, 1600–3. Geography Publications.
Beveridge, E. 1911. North Uist: Its Archaeology and Topography. Edinburgh: Birlinn.
Dixon, T.N. 2005. The Crannogs of Scotland. Stroud: Tempus.
Header Image – Crannog Centre on Loch Tay (Perth & Kinross, Scotland, UK) – Image Credit : PaulT