Terminology topics 4: Atholl

Glen Garry, Atholl

Glen Garry, Atholl, looking north. Photograph © B Keeling

The medieval kingdom of Scotland emerged in the 11th and 12th centuries during the final phases of the Viking Age. Its main sub-divisions were large regional lordships or ‘earldoms’ administered by hereditary rulers who functioned as agents of royal authority. The earldoms succeeded older territorial units which, in some cases, had formerly been small kingdoms in their own right. One earldom of considerable antiquity was Atholl, a region corresponding to the northern part of the modern county of Perthshire. It had once been a Pictish kingdom, with its own line of kings, although these were frequently dominated by powerful neighbours. One king of Atholl suffered a gruesome fate at the hands of the Pictish overklng Óengus, son of Fergus (Unuist, son of Urguist). This was recorded in the the Irish annals under the year 739:

Tolarcan mac Drostan, rex Athfhotla, a bathadh la h-Aengus
‘Talorcan, son of Drostan, king of Athfhotla, was drowned by Óengus’

Here we see an old form of the territorial name Atholl. It appears in other sources as Athfoitle, Athotla or simply Fotla. Historians have long regarded it not as a name of Pictish origin but as a Gaelic name formed by adding the prefix ath, ‘new’ (literally: ‘repeated’) to Fotla, a synonym for Ireland found in medieval Gaelic poetry. Athfhotla is usually interpreted as ‘New Ireland’, a slightly odd name for a part of Pictland, and an imaginative background story has been created to provide a suitable historical context. The conventional explanation goes something like this: Gaelic-speaking migrants from Argyll (i.e. ‘Scots’ from Dál Riata) settled in what is now northern Perthshire sometime before c.700 and renamed it ‘New Ireland’ (because Ireland, like Argyll, was part of Gaeldom). To me, this has always seemed an implausible tale. It seems to be unknown before the late 19th century. Gaelic certainly supplanted the Pictish language but the change involved every Pictish region, not just Atholl, and Ireland didn’t play a major part in the process. Historians usually date the Gaelicisation of the Picts to c.750-900, the period when a lengthy Pictish overlordship of the Scots changed the relationship between the two peoples. The idea that a Gaelic-speaking elite arrived in the centre of Pictland in the 7th century (or earlier) and re-branded a large chunk of Pictish territory as ‘New Ireland’ is unsupported by the sources.

Rather than seeking plausible contexts for an otherwise unrecorded Irish migration to Atholl we can approach the issue in a different way, by wondering if the name Athfhotla might not be Gaelic at all. Could it be a name of Pictish origin?

In his 2009 book From Caledonia to Pictland, James Fraser casts doubt on the ‘New Ireland’ hypothesis. He points out that Athfhotla is not the only name for Atholl in the old sources. Both the Pictish king-list and the Chronicle of the Kings of Alba use variant spellings where the middle element is ochl rather than otl (Achcochlam in the king-list and Athochlach in CKA). Fraser points to the frequent confusion of the letters c and t in early medieval texts and wonders if otl might represent a scribal error for ochl. This mistake could have originated as far back as the 8th century, when the annal entry quoted above was written. Fraser suggests that the correct Gaelic name of Atholl could indeed be preserved by CKA and the king-list, and that it might originally have been Athfochla rather than Athfhotla. This theory makes a lot of sense. Fraser proposes an etymology based on ath-, ‘ford’ (or ‘way through’) and fochla, ‘north’, to give the meaning ‘north way’ or ‘north pass’. He thinks this might be the Gaelic equivalent of a Pictish name which would now be something like Adwy Gogledd in modern Welsh (the living language closest to ancient Pictish). Anyone familiar with the geography of northern Perthshire will appreciate why ‘north pass’ seems such a good description of Atholl. The main feature in the landscape is the corridor formed by the valleys of Strathtummel and Glen Garry, a sinuous route traced by the busy A9 road from Perth to Inverness. Modern travellers soon find themselves weaving through the famous Pass of Killiecrankie before scaling the high Drumochter Pass beside the hill known today as the Sow of Atholl. These people in their cars and trains are simply using one of the main links between northern and southern Pictland, a well-trodden highway through the imposing barrier of the Grampian Mountains. It would be hard to think of a better name for this route than Athfochla, the ancient ‘north pass’ of the Picts.



W.J. Watson, The history of the Celtic place names of Scotland (Edinburgh, 1926), pp.228-9 [for the ‘New Ireland’ etymology]

James Fraser, From Caledonia to Pictland: Scotland to 795 (Edinburgh, 2009), pp.101-2.

* The annal entry for 739 is from the Annals of Tigernach.

6 comments on “Terminology topics 4: Atholl

  1. Pah, and I have such a lovely theory to explain the Gaelic version as well 🙂 Just one of the now circa four thousand reasons I need to get a copy of Fraser’s book this conference season… Thankyou for the warning!

    • Tim says:

      There’s still plenty of room for another Atholl theory, Jon 🙂 And I think you’ll enjoy Fraser’s book….

  2. […] published an interesting post on the possible (probable?) origin and etymology of the place-name Atholl. If you are interested in the indigenous languages of Scotland, then surely you’ll enjoy […]

  3. Tim says:

    Here’s something I wrote today in a comment thread over at the Cumbrian History blog. It’s part of a reply to Jonathan Jarrett and relates to his thoughts on the naming of Atholl (see his comment above). I’m pasting it here because it deals with the ‘New Ireland’ theory which I don’t buy into.

    One major doubt (for me) is why Ireland would be regarded as significant in the naming/renaming of Atholl. With the Irish migration theory of Scottish Gaelic origins now looking increasingly fragile, why would anyone choose to call a Pictish region ‘New Ireland’?

    • Why would anyone name a colony or city… New York, New Jersey, New Orleans or the region New England. Most of the people who lived there, even the immigrants, were not from the city they are named after. I’m sure the native Americans were not wild about the names.

      • Tim says:

        These US colonies were founded by people who had a direct connection with the ‘old’ places: York and Jersey (English settlers); Orleans (French settlers). Also the Dutch and (New) Amsterdam. If Atholl is ‘New Ireland’, then surely this implies large immigration (or, if not, an elite takeover) by people with an Irish connection. At the moment I can’t see what that connection might be.

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