David’s Fort Revisited – and a Strange Coincidence?

By Meryl Marshall

With movements restricted due to the coronavirus pandemic it was inevitable that I would find myself at David’s Fort, near Conan House, just 2kms from my home. This impressive earthwork, variously interpreted as a “motte”, a “moated homestead” (OS map) and a “moated site” has received lots of attention from NOSAS in the past, see Marion Ruscoe’s blog of 2016, but the site remains as mysterious as ever. I was pleased to see that the area is much more open than it used to be, but the surrounds are rapidly becoming overgrown with scrubby brambles, broom and whins. The visit set me thinking once again about the origins and history of the site, with more time at home I set about some online investigations.

David’s Fort (Canmore ID: 12866, Highland Council HER: MHG8986) is at NGR NH 5394 5328 and consists of an impressive wet ditch 4m deep enclosing a trapezoidal area measuring 25m from N to S and 26m to 32m transversely. The ditch is enclosed by an external bank standing up to 3m height but 1.5m externally. Internally the only feature visible is a circular depression 7m in diameter and 1m in depth in the western half; traces of what may have been a bridge spanning the ditch on the west side have also been reported (June 1979) The moat still contains water and was originally fed by a waterway running from an artificially constructed pond possibly of more recent origin 100.0m to the east, to a cut in the bank at the NE corner.

The site is located on the forested slope above the River Conon 1km to the east of Conan House. It is close to what, in the Medieval period, was a crossing of the River Conon. Here too was the old church of Logiebride (or Logie Wester), and the site of the Battle of Lagabraad in 1481. This area, at the “neck” of the Black Isle, will almost certainly have been a meeting point of routeways for centuries, if not millennia.

David’s Fort looking SW

A processed image of Davids Fort from a lidar survey (contains public sector information licensed under the Open Government Licence v3.0) © A Thompson. This model has also been uploaded to Sketchfab and can be seen in 3D at https://sketchfab.com/3d-models/davids-fort-d26cbff5d5184af18d157f7b6be94dad

So what was going on in Ross-shire in the medieval period?

In 1150-51 King David 1st of Scotland (reigned 1124-1153) had campaigned along the Moray Coast, introducing the Anglo-Norman feudal system and granting land to loyal lords who established fortified bases or mottes as far west as Beauly and the Black Isle (the boundary seems to have been around Muir of Ord and the River Conon). King David set up Royal Burghs to encourage trade and established cathedrals and monasteries. But there were some who objected to King David and there was a period of hostility towards the crown in the late 12th to early 13thC. Macdonald (2003) tells us;

The history of the so called Canmore kings in Scotland from Malcolm III (1058 – 93) to Alexander III (1249 – 86) is marked by an array of insurrections led by discontented dynasts and native warlords with grievances against these kings… none of the challenges proved successful… the Canmore kings maintained their grip on power in large measure through crushing rivals and quashing numerous insurrections [these included as early as 1058 the hapless Lulach, stepson and brief successor of Macbeth and later Somerled, Earl Harald Maddadson, the MacHeths (descendents of Gruoch, Macbeths wife) and the MacWilliams (descendents of Duncan (?FitzWilliam) son of King Malcolm III by his first wife Ingjeborg)]… without question the most war-torn regions of Scotland in the late 12th and early 13thC were the large and imperfectly assimilated northern territories of Moray and Ross…

The following quotes from Macdonald (2003) will continue the story:

[p97] One of the mightiest enemies faced by the Scottish kings was without doubt Earl Harald Maddadson of Orkney and Caithness (d1206)… [he] was a dominant figure in northern Scottish politics for over 60 years. [p99] From 1159… he pursued a policy of expansion and consolidation of power in the north … Thus the expedition of King William and his brother Earl David to the north in 1179 … may well have been as much a response to the activities of Earl Harald as to those of the MacWilliams … [p98] and, quite likely… [Earl Harald] was involved in the MacWilliam and MacHeths insurrections that characterise the 1180s [Harald’s wife was a MacHeth] … even though the evidence for this remains problematic.

1179 – William the king of Scotland and David his brother along with the earls and barons of the land went into Ross with a “great and strong army and there they strengthened two castles the name of one being Dunscaith and the name of the other Edradour”.

[p62] The most tenacious opponents of the Scottish kings were the MacWilliams who first appear on the record in 1179/81 and whose opposition was almost endemic down to 1230. [p149] The chronicler Howden relates how Donald MacWilliam landed in 1181 with a large army “wasting and burning as much of the land as he reached putting the folk to flight and slew all he could take”. Howden has a detailed account of how in 1187 Donald MacWilliam was tracked down, quote “using Inverness as a base the royal army set off, 3000 warlike youths under the command of Roland, Lord of Galloway, … approached the army of MacWilliam they made a (surprise) attack upon them and slew MacWilliam himself and many of his army… and the head of MacWilliam they cut off and carried away… and presented it to the king of Scotland” Howden puts the conflict in Ross while the Chronicle of Melrose puts it on a moor called Mam Garvia, probably Strath Garve west of Dingwall also Gesta Analia/Chron Fordun.

[p73] Donalds son, Guthred MacWilliam, continued his father’s cause and is said to have landed in the north from Ireland in 1211. Barnwell annalist tells us “Guthred is of the ancient line of Scottish Kings and supported by the aid of both Scots and Irish, had practised long hostility against the modern kings now in secret now openly as had his father Donald. Bower remarked that this invasion was part of a conspiracy by the nobles of Ross

[p41] Bower tells us “In 1211 William king of Scotland sent a huge army … into Ross against Guthred MacWilliam … on the way he built two castles, laid waste pretty well all of Ross and took or killed many of Guthreds supporters. But Guthred himself avoided the kings’ army laying ambushes whenever he could by night or day and driving off booty. He besieged, captured and destroyed one of the royal castles in Ross demonstrating that he had not yet been decisively defeated.

[p42] But the next year in June 1212 Guthred was betrayed by his followers. He was captured and beheaded.

[p43] King William died in 1214, his activity in the north had not succeeded in quashing opposition and the MacWilliams had not been suppressed.

[p43] In 1215 the MacWilliams and the MacHeths united and entered Moray. This time the leaders were Donald Ban MacWilliam [son of Guthred] and Kenneth MacHeth, together with an Irish Prince [they] joined forces in their opposition to the King of the Scots but were defeated by Ferchar Maccintaigart a native northern magnate.

1215 – Farquhar is made Earl of Ross for his part in the suppression.

Further minor skirmishes from the MacWilliams were to take place in the 1220s but nothing on the same scale and when a MacWilliam daughter was put to death in Forfar in 1230 that was the end of the opposition.

So the conflict and unrest in the area of Ross-shire was considerable and sustained for several decades.

Distribution of moated sites

But to return to David’s Fort; the site is variously described as a motte, moated homestead or moated site. The Canmore definition of a moated site is “a site enclosed within a moat, normally rectangular on plan, and believed to be medieval in date”. But it would seem that “moated sites/homesteads” can also be circular or oval too; nearby Achnasoul earthwork (NGR NH 48808 51821 Canmore ID: 27472), for example, is oval in shape but described as a moated site and there are several others too. Similarly mottes are usually thought of as circular but there are some that are square or rectangular. Mottes, moated homesteads or moated sites? there is some confusion between them but one thing seems certain they are all medieval.

Sticking with moated homesteads/moated sites, research has indicated that they can mark local centres of lordly/?kingly power during the 12th and 13th centuries when Scotland became a feudal society. “Moated homesteads are relatively rare within Scotland as a whole when compared to the frequency of those recorded in other parts of the UK and beyond. There are around 122 known sites in Scotland compared with around 750 in Ireland and 6350 in England…” (quote from Achnasoul Schedule Document). My first investigation was into other moated sites in Scotland, could there be any other sites similar in size, construction and location to David’s Fort? McNeill and McQueen (1996) had a useful map showing a concentration of moated homesteads in the NE and SW of Scotland. This was not really surprising as these are the fertile productive areas, but strangely there were no homesteads around the Edinburgh or Glasgow areas. Two sites, Sir John de Graham’s Castle and Dunrod in Galloway are held up as similar sites to David’s Fort (Waters 2017 and Canmore entry for Sir Johns de Grahams Castle).

Sir John de Graham’s Castle

Sir John de Graham’s Castle (Canmore ID 45283 NGR NS 68132 85849) near Fintry in the Carron Valley, has been interpreted as a square motte, but it is a subrectangular site defended by a broad flat-bottomed ditch, 11m across and 3m deep. The ditch is continuous and access to the castle must have been via a wooden bridge which probably lay on the north-east side. The central platform is almost square and measures 22.8m by 23.4m. To the northeast of the ditch there are traces of a lime-mortared wall and fragments of banks which suggest the positions of ancilliary buildings. Traditionally, the site is thought to have been the residence of Sir John de Graham, who was killed at the battle of Falkirk in 1298, but the site may be of earlier date and probably the principal stronghold of the barony of Dundaff which is on record in 1237.

Dunrod moated site

Dunrod moated site (Canmore ID 63927 NGR NX 6997 4591) SE of Kircudbright in Dumfires-shire. This homestead moat comprises a rectangular platform bounded by a fragmentary stony bank and a substantial ditch. It is larger than Davids Fort, the interior platform measuring 37.9m by 34m. The surrounding moat is large and on the SW rock cut, measuring 8m in width and up to 1.8m in depth.

Initially I explored other site on Canmore in Aberdeenshire but many seemed poorly preserved or had little online information. So I looked further afield. I was aware from my many visits to Glasgow that there was a site called Peel of Gartfarren near Aberfoyle. I had seen it marked on the OS map and we had passed close by many times but never visited. To my surprise the Canmore website had not one but two well preserved moated homesteads in that area and even more remarkable was that they were both trapezoidal in shape and similar in size to Davids Fort. The difference was that they were in low lying marshy ground (Flanders Moss), hardly a landscape similar to Davids Fort.

Peel Of Gartfarren (Drymen Parish) Canmore ID 44640, NGR NS 5361 9537 (Quote from Canmore) “the Peel of Gartfarren is one of the best-preserved and most accessible examples of a moated homestead to be found in central Scotland. Like its near neighbour Ballangrew, it is trapezoidal on plan and defended by a relatively low inner rampart with a characteristic broad, flat-bottomed ditch accompanied by a low outer bank. Although the ditch is now dry, it may originally have been filled with water, and there are faint traces of what may be a feeder channel leading from a small stream to the south. The entrance is on the west, facing the road, and is marked by a gap in the rampart and a corresponding causeway across the ditch. The south section of the rampart has been extensively robbed, and on the south-west there are clear traces of where the carts have been brought into the site to carry away the stones. On the other three sides of the enclosure, there is a capping of stones on the crest of the rampart, suggesting that at some time a wall had been built on top of it, but it is not clear whether this is an original feature. In the north-east angle of the interior, there are the foundations of a rectangular building which is probably of comparatively recent date as the medieval buildings are more likely to have been of timber”.

Peel Of Gartfarren

Ballangrew (Flanders Moss/Lake of Mentieth) Canmore ID 45371 NGR NS 61774 98868 (Quote from Canmore) “the well-preserved moated site at Ballangrew now lies on the north-west margin of Flanders Moss, close to agricultural land, but in medieval times, before a phase of extensive land reclamation in the 18th and early 19th centuries, it probably stood deep in the confines of the Moss and may have served as a hunting lodge. Trapezoidal on plan, the central platform measures 23m by 21m within a broad water-filled ditch up to 8m across which from the outset was probably intended to be wet. There is a slight indication of an inner bank on the south, but the bulk of the material from the ditch has been placed on its outer lip to create a bank up to 1m in height. Access to the interior must have been via a wooden bridge as there is no sign of a causeway across the ditch.”


My next port of call was the name “David” – who was the person who had given his name to the site? The OS Name Book of 1876 has: “The origin of the name ‘David’s Fort’ is unknown” – not much help there! Marion in her blog suggests that the site may have been built at the behest of David, Earl of Huntingdon, brother to King William the Lion – as we have seen William and David were campaigning in this area in 1179-81 and had “strengthened” the castles at Dunscaith and Eddradour – was David also responsible for Davids Fort at that time or perhaps even earlier. This was a strong possibility; my next mission was to find out more about David of Huntingdon.

David, Earl of Huntingdon (1152 – 1219) was an influential Scottish Prince and the younger brother of both King Malcolm IV (1153 – 1165) and King William the Lion (1165-1214). His paternal grandfather was King David I (1124 – 1153) who was described as one of Scotlands greatest medieval kings. The period of 60 years, during the reigns of both Malcolm and William, following the death of King David, were a time of great unrest and campaigning. Not only was there trouble from the several claimants to the crown, but the border between Scotland and England was hotly contested. Malcolm IV had succeeded to the throne at the age of 11 in 1153 and four years later he ceded the counties of Northumbria, Cumberland and Westmoreland to King Henry II of England in exchange for the Earldom of Huntingdon. His brother William (the heir to the throne) was captured during an attempt to regain the counties but was released in 1163 in exchange for David (then aged 11) who was sent to the English court as a hostage. David returned to Scotland as heir apparent when Malcolm died in 1165 and William succeeded to the Scottish throne. But William invaded Northumberland again attacking Carlisle and Alnwick; he was captured by King Henry II of England and not released until 1174 when he agreed the “treaty of Falaise”. David, who also returned to Scotland, was to play a useful part in supporting William after 1174, he was created Earl of Lennox and granted the district of Garioch, in Aberdeenshire. The Earldom of Huntingdon, which had been confiscated in 1174 (?as part of the Treaty of Falaise), was restored to William the Lion and transferred to David in 1185.

David had other notable achievements too:

  • He carried one of the three swords at the Coronation of Richard the Lionheart on 3 September 1189 and is said to have accompanied King Richard on the third crusade to the Holy Land in 1189-1192
  • A possible Robin Hood connection – David is a possible inspiration figure for the Robin Hood legend because the legend plays in the 1190s at the same time as David. The association of Robin Hood with the Earl of Huntingdon can be traced to ballads of the 17th century, such as A True Tale of Robin Hood. Both David and Robin Hood are said to have taken part in the Third Crusade, and by 1194 David had taken part at the siege of Nottingham Castle where the High Sheriff of Nottinghamshire and Derby County was taken captive.
  • Last but not least the two main claimants in the line of succession to the Scottish throne on the death of Margaret, Maid of Norway in 1290 (and Alexander III in 1286) came from David, Earl of Huntingdons’ descendants – those of Baliol and Bruce. Earl David had seven children, the two who were most involved with the succession being:
    – Margaret of Huntingdon (the eldest daughter) who married Alan, Lord of Galloway, they had two daughters – one of which was Devorguilla of Galloway, she married John de Baliol and their son became King John Baliol (crowned 1292).
    – Isobel of Huntingdon (the youngest daughter) who married Robert Bruce, 4th Lord of Annandale, their grandson Robert, Earl of Carrick, became the famous King Robert the Bruce.

David was indeed a noble figure. But where did moated homesteads and Davids Fort fit in? I decided to follow up the Lennox connection, largely because I knew Lennox was north of Glasgow (the training ground of Celtic Football Club is at Lennoxtown!) and I suspected that the two moated homesteads of Gartfarran and Ballangrew might be in the Earldom of Lennox.
The old Celtic Earldom of Lennox no longer exists; it covered what today is Dunbartonshire, a large part of Stirlingshire, and parts of Perthshire and Renfrewshire. Following the medieval period its boundaries seem to have changed significantly over the centuries.

The following is an extract taken from a summary of a talk “The origins of the Earldom of Lennox” given to the Centre for Scottish and Celtic Studies in 2012, by Matthew Hammond; http://cscs.academicblogs.co.uk/the-origins-of-the-earldom-of-lennox/

In the 12th century, the earldom of Lennox was not part of Scotland as we know it now. In Di Situ Albanie, composed between 1165 and 1184, the bounds of Scotland stretch from Caithness and Ross, included Argyll but not this area of Lennox, just north of Glasgow. At the time, warrantors from Cowal and Kintyre were to correspond with the earl of Menteith, essentially by-passing the earl of Lennox.Lennox was very much part of the Irish Sea zone, a Hiberno/Norse context. It was an expansive region, Gaelic-speaking and had several links to St. Kentigern and St. Patrick. Another factor that made it a distinct entity in Scotland was the use of a unique unit of land assessment. It also belonged to the diocese of Glasgow…..it seems to have been part of the kingdom of Cumbria, revived by David I. David, earl of Huntingdon (grandson of David I) was given Lennox by his elder brother, King William, in 1174. It is first described as an earldom in a charter from 1178. … Indeed, David controlled the lordship for 11 years, until it was relinquished in 1185, after he was given Huntingdon. Dr. Hammond asserted that the creation of the earldom of Lennox may have been down to David’s appointment and it was raised up from lordship to earldom on his account.

Stringer has also argued that Lennox was a strategically important area in Scotland and control of it would allow consolidation of royal power. Other historians have argued that it would merely act as a buffer zone, sealing off the threat of Somerled and his descendants. [Summary by Ross Crawford, PhD Researcher].

Flanders Moss and the Carron valley formed the northern edge of the Earldom of Lennox. The three moated sites of Peel of Gartfarren, Ballangrew and Sir John de Grahams Castle are close to that boundary and would appear to be “guarding” it in a similar manner to that of Davids Fort guarding the river crossing in Ross-shire. The possibility of the sites being “hunting lodges” has been suggested and must not be ruled out either, the landscapes in all cases were probably ideal terrain for hunting. What is remarkable about the four sites is their similarity in size and shape, particularly the unusual trapezoidal shape (although Sir John de Grahams Castle less so) – why trapezoidal? There must have been a reason for it! The possibility that the four sites are the work of one “architect” is not to be ignored and maybe that person was David, Earl of Lennox (later Earl of Huntingdon). As we have seen from the historical research he had connections with the locations of all of the sites. David, Earl of Huntingdon, surely is the prime candidate responsible for David’s Fort at Conan.

Or is this all just a strange coincidence?


Grant Alerxander 2005, The Province of Ross and the Kingdom of Alba (From “Alba, Celtic Scotland in the Medieval Era edited Cowan and Macdonald)
Macdonald R Andrew 2003, Outlaws of Medieval Scotland – Challenges to the Canmore Kings 1058 -1266
Marsden John 2010, Kings Mormaers Rebels – Early Scotlands other Royal Family
McNeil and McQueen 1996 Atlas of Scottish History to 1707
Oram Richard 2011, Domination and Lordship Scotland 1070 – 1230,
Waters Geoff 2017, Field Excursion Guide to Black Isle and Glen Urquhart (AS Summer School 2017)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.