A Survey of the Loch Awe Crannogs

Team Leader and Organiser: Marcia Taylor

Team Leader: Mark Holley

Participants: Ken Washburn

No part of this report text is to be reproduced without the express permission of the author-
M. Taylor T/L 41 Apsely Street, Partick, Glasgow, G11 7SN.

In 1999 financial support was generously given by the British Sub-Aqua Club – Jubilee Trust (BSAC) to aid the underwater survey of submerged archaeological features known as crannogs in Loch Awe, Argyll. The project took place in June 1999 and involved the three participants detailed above. The project remit was to re-investigate and acquire detailed, accurate measurements (to the current modern standard) of those sites determined to be Crannog settlements during a survey of the loch in 1973 . Furthermore it was undertaken to obtain suitable material where possible for the determination of radiocarbon dates for the structures.

Arrangements were made by telephone and letter to keep the locals, landowners, forestry and wildlife officers informed about the presence of the survey team and project and their wishes were respected.

  1. To survey all submerged settlement sites, known and as yet undiscovered in Loch Awe using remote sensing and long distance measurement equipment.
  2. Assess individual site condition and stability due to impact of leisure or other activities on the Loch.
  3. To retrieve any suitable samples for Radiocarbon dating.
  4. To address the potential historical use of the artificial islands by detailed recording of any underwater features or superstructures.
  5. To record the sites using underwater photography.
  6. To publish survey and dating results in an appropriate academic journals and popular or specialised press.
  7. The deposition of documentary and photographic archive relating to the project with the correct government bodies after project completion.

The history of research into artificial structures in lochs started back in 1863 when the Loch of the Clans was drained. Two stone mounds were uncovered that contained timber piles and rafters . From then on sites are occasionally mentioned and reports concerning crannog structures appear with increasing frequency in the Proceedings of the Society of Antiquaries. However it was not until 1879 that the first serious research into to these structures began. Robert Munro, a 53-year-old doctor from Ayrshire had an amateur interest in archaeology. He worked in the south west of Scotland compiling data from agricultural drainage sites that had reported crannog mounds. Hence when he published the first account his “Ancient Scottish Lake Dwellings or Crannogs” in 1882 it was based mainly on information he had painstakingly collected from museums, the knowledge of local farmers and his own first hand experience of excavation . This book is still valued today for its clinical mechanical approach to information about the sites excavated and it is still the largest collection of survey work yet published.

Munro excavated the drained site of Buston Crannog. The report is exceptional for its period in that it included measured and accurate plans of the findings. Interlacing wooden joints, hearths, midden areas and a well-preserved canoe were all recorded. Because of the report detail Buston quickly became the basis for the public notion of the construction of crannogs . The discoveries at the Buston excavation were reflected shortly after in the information from J. Stewart’s investigations at the drained site of Dowalton Loch. Here six crannogs were discovered as well as a canoe or boat structure and evidence of morticed timbers jointed timbers, pegs and hearths of stone and clay. Thus the stereotypical image of crannog dwellings was born: as wooden pile structures supporting a domestic dwelling place accessed by causeways from the shore and boats from the open loch.

The Reverend Otto Blundell was the other main early antiquarian who in contrast to Munro worked principally in the highlands. Blundell was amongst the very first archaeologists in Britain to use diving techniques in person. He borrowed equipment from the Caledonian Canal divers (one of the earliest Sub Aqua groups in Scotland) and went under the water himself to observe the undrained structures in a number of Lochs, such as Eilean Muireach or Cherry Island in Loch Ness in 1908 . Being a monk he used this influence with all the monasteries and churches across Scotland to send out a questionnaire asking all monks to find crannogs in their local area. Using this survey he enlarged the known database by 500 sites. This moved the emphasis from the SW and made people realise the countrywide distribution of these sites.

The advent of the First World War produced a hiatus in the investigated of the crannog mounds and this situation continued to be during the first fifty years of the 20th century. Thus the stereotype crannog images were not challenged and became established as generic archaeological fact. This situation did not improve greatly upon the 1953 excavation of a mound in Milton Loch, as the site structurally resembled Munro’s earlier excavation sites and was also located in the south-west area of Scotland . A variety of reconstruction images followed the Milton excavations and became the standard image of all crannog types in published literature. This imagery has continued to be re-enforced through full-scale replication of pile structures in Craggaunowen, Co Clare, Ireland in the 1970’s and at Loch Tay in the 1990’s.

During the 1970-80’s a change in Archaeological thinking in general saw crannog mounds taken out of isolation and considered in equality with land sites and the landscape as a whole. There was a shift from regarding the sites themselves as important and a move towards general landscape studies. The application of survey strategies as opposed to excavations enabled this shift in emphasis away from individual sites and onto the interplay of crannogs with their landscape. Crannog sites became components in whole systems of cultural, economic and environmental interactions.

The first wide-scale survey took place in Loch Awe in 1973 . 60 possible crannog sites were defined from this one Loch from map work and research. A team of Royal Navy divers helped the academics to establish 20 definite crannogs out of the 60 in 3 weeks. The Loch Awe survey became groundbreaking in that it established the pattern of survey aims and methodologies for all future work. The 1973 data achieved also became the foundations for accepted notions of size, crannog type and expected features (see below for critique).

The Loch Awe survey was followed by another survey of an inland water body, Loch Tay in 1982 . This survey was conducted in a similar manner with the same basic academic agendas, although with more modern recording equipment and techniques. The information from this work resulted in the first modern excavation of a fully submerged crannog and it subsequently lead to the erection of a full-scale reproduction following a possible interpretation of the original excavated structure, Oakbank.

In the 1990’s three further surveys and two trial excavations of crannogs have taken place in Scottish waters. Of the two surveys, one was conducted within Scotland’s only inland freshwater Lake, the Lake of Menteith and the other took place on the islands of the Inner Hebrides . All freshwater bodies on these islands were surveyed and the crannogs duly noted. Dating material was recovered where possible and has provided a series of new dates for an area of the country that has long been over looked . Investigations have begun into the presence of inter-tidal crannogs in Scottish coastal waters. Sites have been noted and trial exploratory excavations have taken place at Redcastle, Phopacy, Carn Dubh and Erskine Bridge in the Moray Firth and the along banks of the Clyde at Dumbuck.

When reviewing the primary Loch Awe survey data-set in advance of the Inner Hebrides study project, the team leaders found a host of data inaccuracies which lead them to suspect limitations of the 1970s material. Over the thirty years 8 publications have used, quoted or discussed the data with varying levels of accuracy. What is significant about these sources is that there are major inconsistencies in the data reported within and between the various publications.

Inconsistencies have always been present in the Loch Awe data. Individual site reports were not included in the original S.A.C. publication, the site measurements between the illustrations and those reported in the text do not concur and there are fundamental mistakes over the use of metric and non-metric scales. Indeed the survey data is presented as summarised findings with hand drawn plans for only six of the sites . In the report timber was identified from nine of the sites but only specifically mentioned in the text on six sites. M Holley has discovered that all of the measurements taken are rounded to the nearest metre or foot and in some cases are estimates. In the report minor discrepancies between the crannog site plans were not considered to be important as they were, “only meant to give a rough idea of the shape of the sites” (D. McArdle pers. comm. to M. Holley 1997). None of this is surprising in that the 1972 survey agenda was only to demonstrate that crannog sites were actually present in the loch. The plans produced were only meant to illustrate the shapes of the sites. Numbers of points taken on particular sites varied. In some cases the outline of the crannog base was produced from as few as 5 measured points, Inistrynich, Rockhill and Eredine . Further publications took no account of any of these difficulties and therefore more confusion arose through the replication of mistakes and inaccuracies. Indeed many of the subsequent publications introduced their own measurement or graphic errors.

It became obvious that a re-survey of the loch was necessary to iron out the compounded difficulties that had encroached upon the data since the original publication in 1973.

The survey team consisted of three members. Their individual roles were defined as Diver, surface support snorkel diver and remote Electronic Distance Measurer (EDM) operator. Access to the sites was achieved mostly by boat transport. On occasion some sites could be reached from the roadside. Electronic survey equipment was set up on shore whilst the diving party prepared by conducting an initial swim-over site assessment. The survey was then conducted with two fully kited divers in the water and an EDM operative on shore. Survey equipment consisted of a Lica EDM and laser reflector prism on an adjustable pole. Surveying using an EDM machine provides high quality accurate measurements and has the added advantage that these can be achieved over distances of up to 1 kilometre with ease. One diver descended below the water to be able to place the end of the prism pole on the points to be measured, whilst the remaining diver stayed afloat at the surface to orientate the prism and serve to communicate between the submerged diver and the shore operator. On average 150-200 points are taken per site. This figure can vary dependent upon the crannog size and number of features present. The survey strategy began by recording shore outlines where appropriate and then attention turned to the crannog site. On each site an east-west and a north-south transect was recorded following compass point readings between surveyors ranging poles. These transects gave the site profile. Then the base outline was recorded along with any upper platform outline if existing. After this the individual features of the site were recorded. Natural aspects or features of the site were outlined such as the location of bedrock protrusions. The location of anthropogenic timbers or other structural features were also mapped to be added to the eventual complete survey plan.

Sites were recorded at the rate of 3 to 4 a day in order to complete the project within the available time. Whilst in or passing appropriate locations along the loch any anomalies or other suspected crannog sites were investigated. During survey visits at the sites the state of the structure’s stability was noted as were any recent signs of disturbance or alterations that could provide future problems with the escalation of decay at the site. All the known sites are scheduled ancient monuments and thus are protected from interference by the government body Historic Scotland.





Timber reported

Associated Reported StructuresNotes from 1999 Survey
Larach BanNN02SE18



Structural timber sighted through boulder mound
Eiliean SeileachanNN02SE19


JettyNo Jetty noted
Inishail ChurchNN02SE20


Harbour FeatureNo Harbour Feature


Upper level raised in 19thcentury with stone capping



Possible field clearance cairn Heavily silted and obscured No secure structural timbers noted


JettyNo Jetty


  1.  Eastern end of Loch Awe in the vicinity of Kilchurn castle.
  2. Area of outflow of River Awe in to the Pass of Brander.
  3. Ardanaiseig point
  4. Area between Ederline and Fincharn Castle

To survey all submerged settlement sites, known and as yet undiscovered in Loch Awe using remote sensing and long distance measurement equipment.

This task involved the complete re-survey of all the 20 known sites stretching the length of Loch Awe.

The sites were located through a combination of reference to the original archive data from the 1972 project, the RCAHMS published inventory list and the limited use of an echo sounder . Each site was recorded in detail through the use of a Lica EDM and observation field notes were taken pertinent to the state of the individual site. The twenty sites corresponded roughly to their published locations although there were inevitably some fluctuations between actual distances from shore and the estimations published in the RCAHMS volumes. This may be explained by two factors.



  1. Any estimate of distance is personal and subject to the accuracy of the viewing eye. The RCAHMS report relies upon this for all the sites with visual above water portions as no offshore visits were made to the crannogs by the inspectors. The submerged site estimates came from the original 1972 survey data.
  2. Electronically measured distances although more precise than those produced by the eye are also subject to variation. The fluctuation of the water levels in the Loch on a day-to-day and seasonal basis will effect the physical distance of site from shore. Thus any measurement of distance should be regarded as a fluid statistic dependent upon raised or lowered loch levels.



Three sites Keppochan, Inistrynich and Larach Ban show significant changes and some similarities, as would be expected, between the 1999 and the 1972 plans. The site recorded at Larach Ban shows particular differences in outline and confirms the fact that some of the 1972 plans which utilised only 5 measured points did produce misleading records.

Areas Investigated


The use of echo sounding equipment on the project was limited. Funds were not available for the hire of this machinery for the duration of the project. Thus it was used with effect only in one area of the Loch. The eastern end of the loch in the area around Kilchurn Castle was investigated. Here there are a number of rocky protrusions from the loch bed that could have been amplified and exploited as crannog sites in the past. No trace of such structures was found suggesting that all evidence of earlier dwellings had been removed or that they did not exist in this area.

Prior to the commencement of the project desk based research produced several other areas that had the potential to contain crannog structures or had reported suspect crannogs from folk history. Exploratory investigation proceeded without the aid of electronic machinery. A swim over survey method was adopted using three snorkelling divers at spaced intervals to check the particular areas of the loch. The areas investigated detailed above did not reveal any other likely sites or anomalies. Area’s 1, 4 and 6 contained suspected crannogs. In areas 1 and 4 no relevant anomaly bodies were located.

Access was denied at Area 6 due to its current use as recreational fishing and yachting tasks.

“A roughly circular setting of stones, normally under water, was revealed towards the end of June by the abnormally low level of Loch Awe. Investigated by boat and by wading, the setting of small boulders was about 7m in diameter, with another, smaller approximately rectangular setting of boulders slightly off-centre about 2m by 1m within. At the time of visiting, it stood in about 10cm of water, on a shallow mound about 15m in diameter separated from the shore by a channel between 1m and 2m in depth.”

It is suspected from the description quoted above in the RCAHMS inventory (accessed online November1999) that the site may have similarities to that at Inistrynich and may be the result of the submergence of agricultural clearance cairns. As with Inistrynich this hypothesis remains to be explored more fully.

The area around Ardanaiseig point, area 3 was thoroughly investigated as local knowledge and history from a resident indicated there might be structures both on land and below water. This search proved to be fruitless in both mediums.

One other area of note discovered during the course of the survey exists. Directly behind and slightly east of the Carn an Roin site a structure was discovered. It does not exist in any of the known site inventories for the region and thus may be a new unidentified archaeological structure. It currently lies on the shore of the loch within the deciduous, shore lining trees and shrubs. The site consists of a 2-3 metre high boulder mound, with an oval / circular layout. Raising the loch level by as little as 2 metres would flood this site. Tree and shrub roots and a build up soil covered the mound. This indicates that the structure has not been submerged for a considerable period if indeed it ever was. Visually the construction of the boulder mound was highly similar to the nearby crannogs. It is possible that it may be the remains such, but there were also features that could lead to alternative interpretations. The mound exhibited a possible kerbing around the exterior and the sizes of individual boulders did not correspond in all areas to those usually found on crannogs. Thus it is pure speculation to define the site as an exposed crannog from a visual inspection alone. The site could be an ancient burial cairn although this would be extremely close to the water if not in it at times. Or the mound may have been produced by the clearance of fields. Whichever explanation is correct the site definitely warrants further exploration (out-with the scope of this project) to determine its nature.

  Assess individual site condition and stability due to impact of leisure or other activities on the Loch.

The sites of Sonachan, Ederline Boathouse, Inverliever, Carn an Roin, Larach Ban, Keppochan, Carn Dubh and Rockhill were identified as being the main sites impacted upon by current human activities at the Loch.

The main activity taking place over the whole extent of the Loch is related to fishing. The crannog sites are natural attractions as habitats for the fish. This produces a double-edged sword situation as knowledge from fishermen and anglers can provide information of site location for the archaeologist. However the attraction of the fish to the sites and subsequently the anglers can also have detrimental effect, with the creation of modern boat noosts or other modifications to the site. It is undoubted that the loch supports a large seasonal fishing population, which has some impact upon the seclusion of these sites. Just after the completion of the project an unfortunate incident with a hired fishing boat occurred which resulted in an accident with fatalities for those concerned.

However from close observation of the sites apart from the deposition of quantities of lost fishing line or weights (more hazardous to divers) there appears to be little or no serious damage occurring to the sites. The majority of sites with surface areas above the water are not large enough to induce frequent landings and most that do come above water are more isolated from shore. The exceptions being Ederline boathouse and Inverliever at the west end of the Loch. These sites are exposed and located within easy reach of swimmers or recreational holidaymakers, particularly in the case of Inverliever. However aside from the build up of rubbish and aluminium cans at the Inverliever site there seems to be little lasting damage. Although this should not encourage complacency as the accumulation of visual human lifestyle waste can condition the sites by revealing them as visited or visitable as well as spoiling the scenic beauty of these places. Thus it was generally concluded that fishing and boat hire on the Loch currently rests within stable and low impact conflict with the archaeological sites.

The visitor attraction to the area from the Ben Cruachan power station and visitor numbers generated by the Forestry commission managed Tourist picnic sites, seem also to have little if not no impact. The crannog sites are mostly hidden and unemphasised features of the modern leisure landscape around Loch Awe. Even at the one site Carn Dubh that has been incorporated into the Forestry leisure management there is no sign of disturbance despite the fact this site is a mere 10 metres from the shore and is signposted.

In fact only one serious adjustment to a site was recorded by this survey and that was not due to any impact thorough the leisure industry. Sonachan crannog displays serious modification through the robbing and removal of the upper boulder capping mass. The interference has been caused by the adjacent farm and shore landowners attempting to improve their land for cattle grazing. The landholders have tried to reclaim land at the shore off which the crannog lies. To do this they have begun to strip the upper crannog levels, exposing timbers and lower boulder mass. This has undoubtedly resulted in the degradation and decay of the site and has altered its appearance irreparably. The cost of this action to the ancient monument is incalculable and may have resulted in incurable disturbances to the site equilibrium. As the crannog is a scheduled ancient monument, proceedings have been taken out against the individuals concerned by Historic Scotland on behalf of the government and the Scottish nation to raise a fine for damages incurred and prevent further future action. For the purposes of this survey the damaging action has one positive side, despite hostile reception, the team were able to obtain a timber sample for dating purposes.

 To retrieve any suitable samples for Radiocarbon dating.

Timber was discovered during the survey on several of the crannogs in the Loch. However suitable samples were only retrievable from four sites, Ederline Boathouse, Sonachan, Carn an Roin, Keppochan. As the sites are scheduled ancient monuments no disturbance of them is to be tolerated. Thus samples were only taken where timbers were retrieved without moving structural boulders on the sites. At this time the dating samples are under analysis (as separate funds had to be raised for the cost of the dating process) and the results cannot be included in this report. Preliminary indications are that the sites will date to the prehistoric Iron Age period. This tallies well with the only other date for Loch Awe which was achieved from the Ederline Boathouse site. A timber from this crannog produced a date of 370± 45 bc (UB-2415) or 2220± 45 bp (GU-2415) this places it within the time frame of 400BC- 190BC, or the later half of the Iron Age.

In a recent journal publication the topic the chronology of crannog utilisation has been addressed. In 1998 Henderson compiled a review of all the known dates for Scottish sites . The data patterns produced infers that the commonly held perception of the longevity of crannog use from the earliest Neolithic through to the later medieval period may be misleading. He argues that most dates cluster around the later half of the Iron Age and thus show a “crannog building boom”, which is really the representative period for this type of prehistoric structure. This may well be true but a word of caution should be exercised. Datable timbers very often are retrieved from the exterior, the top or the midst of the structure through the boulder mass. They do not come from the middle of the base of the central mound and thus may not be truly representative of the earliest phase of the crannog on the site. Also the Iron age boom period may be over emphasised for the inland and mainland lochs as there are less instances of structural timber use in the islands due to availability of natural resources and thus less opportunity for dating material. And when the whole picture is looked at the reality of the situation is that currently we have a real brevity of dates in comparison to the number of actual crannog sites known.

  To address the potential historical use of the artificial islands by detailed recording of any underwater features or superstructures.

The only traces of historical period use of the crannog sites were activities that could be attributed to the 19th century. On Ardanaiseig and Ederline boathouse crannogs the sites were both modified. Ardanaiseig by the addition of boulders to raise the height of the crannog to project out of the water. This provided a focal and fashionable talking point in the bay in front of the estate house for the lodge guests and Edwardian upper class during the early days of archaeological antiquarians.

At Ederline the removal of the upper layers of the boulder mass in a similar manner to that described at Sonachan took place. Ederline boathouse may also have been modified during the earlier past however it is safer to assign the main boulder removal to later day activity which is documented. The site at Eredine is part of an estate, managed and developed during the Victorian and Edwardian eras. Upon exploration of the crannog site it was noted that the extent and height of this site was unusually small in both respects. The site’s closeness to shore also lead to the suspicion that this site may originate in the 19th century development of the estate lands. A precedent for this type of activity was set during the Victorian period when the crannog at Spar Island in Loch Tay was extended and modified for a royal picnic for Queen Victoria during her highland travels. Refurbishment of the crannog took place was for the royal parties visit. There is evidence in Ireland for certain occupation of sites in the 19th century and it is not at all unlikely that the same was happening in areas of Scotland. Just prior to this there are Historical references which allude to the use of built up islands during the 1745 retreats as 18th century prisons and they were heavily used after the introduction of whiskey duty for hide outs and illicit stills. Thus we cannot reliably know when chronologically crannogs fell out of use.

  To record the sites using underwater photography.

It was not possible to create a photographic archive due to the poor visibility of the water within the loch during the dates the project took place. The shots that were taken were marred either by slits kicked into suspension by the divers or were too dark.

  To publish survey and dating results in an appropriate academic journals and popular or specialised press i.e. Proceedings of the Society of Antiquaries of Scotland, Diver Magazine.

This will realistically occur later this year. When all the final information is in place, the plans collected in one place and the timber samples sent for analysis return results, then the final report can be prepared. A copy of this will be lodged with the BSAC Jubilee trust when it is completed.

  The deposition of documentary and photographic archive relating to the project with the correct government bodies after project completion i.e. Royal Commission on Ancient and Historic Monuments for Scotland

The results of the re-survey project of the ancient crannog structures preserved in Loch Awe have generated a wealth of information. Significant differences were recognised from the present survey data compared to the data set recovered from the 1972 survey. Principally there are disagreements to the precise size and shape of many of the crannogs. The 1999 re-survey was conducted using scientifically precise measuring instruments that rely upon laser technology to achieve measurements, with the ability to do this over long distances. The 1999 survey will eventually produce a full and complete set of detailed plans for each site, encompassing height, depth, distance from modern shore levels, profiles and outlines. This should clear up once and for all the compounded measurement errors and structural data confusions that exist within the published archaeological literature dealing with the Loch Awe crannogs since their original publication in 1973.

Discrepancies exist between the presence and absence of structural elements noted to occur at certain sites from each survey. Often harbour features or jetty structures were reported from the original diving work in the loch. Upon re-examination of these features many of them were found to be no more than areas of the main crannog boulder body that had coincidental visual similarities to real structural elements. The misrepresentation is clearly understandable, as many of the divers who conducted the original survey had no archaeological training (D. McArdle pers.com. to M Holley 1997).

Doubt was cast on the authenticity of two of the twenty originally recognised sites, Eredine and Inistrynich. Although without exploratory trial trenching and excavation of these mounds it can not be said with certainty that they are not crannogs. A third previously unknown site was discovered on dry land, near to a collection of sites at the eastern end of the Loch. This structure would again require further testing to establish its archaeological nature. No other new submerged archaeological sites were discovered.

The general quality of the preservation of the sites was discovered to be good. There had been little interference to the sites from recent modern leisure or other activity-related pursuits over the last 30 years. However the crannog at Sonachan proved the exception to the rule. Here the current owners of the adjacent shoreline had removed a large number of the upper boulders of the site in an attempt to reclaim grazing land from the loch bed. As all the sites are protected under the scheduled ancient monuments act, the government body Historic Scotland are pursuing legal fine procedures to prevent further damage to the site.

A series of radiocarbon dates for several of the crannog sites is under production at an American laboratory. Current early indications are that the dates fall into a range that puts the sites into the later half of the prehistoric period during the Iron Age. This corresponds well with the single date known from Loch Awe and in general with the dates produced from other mainland freshwater Loch bodies. The timber samples recovered from the crannogs were taken from the upper and mid-levels of the sites and it must be stressed that the dates produced may reflect only one phase of occupation at the site and not its original structural beginnings. To gain more detailed information of this type costly and long running excavation of the crannog mounds would be required. This lies out-with the scope if this survey project.

The 1999 survey project has provided invaluable, clear and concise data regarding the ancient crannog structures that exist in Loch Awe. The information acquired from this survey has greatly enhanced the physical knowledge of these sites and their period of use. The completion of this re-survey project has benefited current archaeological research in underwater archaeology immensely and Scottish archaeology has also gained in the wider context from the extra knowledge and precision of data produced from this project.

The team would like to thank Ian W Morrison, Department of Archaeology who on behalf of the University of Edinburgh, arranged the use of a Leica EDM for the duration of the project. Ian Macdonald is thanked for his wit and local wisdom which made the hire of the boat a pleasant everyday experience. Dr McAardle and Dr Ian A Morrison, Department of Geography, University of Edinburgh are thanked for their support and helpful attitude towards the reassessment of their earlier pioneering work. Especially the team would like to thank the BSAC Jubilee Trust for the provision of funds towards essential accommodation and subsistence and fuel costs without which the project would not have been possible.

© 2016 crannogs.com