Déisi · Dyfed · Pembrokeshire · Killeen Cormac
A displaced Irish tribe crossed the sea, founded a kingdom, and left behind the most unusual Ogham inscriptions in the entire corpus — stones that speak Irish and Latin simultaneously, to two audiences at once.
The first thing to understand
One of the most immediately striking facts about Ogham outside Ireland is how differently the script behaves the moment it crosses the Irish Sea. In Ireland, the formula is almost invariable: Ogham alone, on the edge of a stone, recording a name and a lineage in Primitive Irish. No Latin. No second language. The inscription is complete in itself.
In Wales, everything changes. Many Welsh Ogham stones are bilingual — text inscribed vertically in Ogham on the edge of the stone, and the equivalent horizontally in Latin script on the face. The same information presented twice, in two scripts, in two languages, on two different surfaces of the same object. This is not a technical variation. It is a physical record of people living between two worlds — people who needed to speak to an Irish audience and a Roman one, and chose to do both, on the same stone.
The consistent absence of Latin from Irish stones, and its consistent presence on the British stones, is one of the most eloquent things the corpus tells us. In Ireland, Ogham was sufficient and self-contained. In Wales, it needed a partner.
These inscriptions in Ogham indicate a desire to elevate Old Irish language to the same level and status as Latin. The bilingual stone is not just a practical convenience — it is a political statement.
Thomas Charles-Edwards, Early Christian Ireland (2000), pp. 176–177Who made these stones?
The distribution of Ogham stones in Wales is not random. It maps almost precisely onto the territory of a single Irish tribal group whose story is one of the most remarkable in the post-Roman Atlantic world.
The Déisi — the name originally meant "client people" or "landless vassals" — were expelled from the royal seat of Tara as a result of a violent rupture with Cormac mac Airt, High King of Ireland, at some point in the third century AD. The legend of their expulsion is vivid and specific: it begins with Oengus, a prince of the Déisi, killing the son of Cormac in revenge for the rape of his niece. In the fight, he also blinded Cormac in one eye, disqualifying him from the kingship. Whether this account is history, myth, or an elaborate origin story attached to a migration that happened for other reasons, it served as the founding narrative of one of the most archaeologically legible population movements of the early medieval period.
Expelled from Tara, the Déisi settled across central and southern Ireland, eventually colonising the Waterford and Dungarvan area. That territory still bears their name — Waterford's hinterland is still called "the Decies," from Déisi. The old railway greenway between Dungarvan and Waterford is the Rian na nDéisi — the Decies Trail. And it was from this coast, in the late 4th and early 5th century, that large numbers of the Déisi crossed the Irish Sea to Britain.
They settled in the land of the Demetae in south-west Wales, displaced the original British ruling class, and founded the Kingdom of Dyfed — bounded on the north by the River Teifi and on the east by the River Tywi. It may seem puzzling that there are more Ogham stones in Pembrokeshire than in Wexford, the Irish county immediately across the water. The answer is that it was not Wexford but Waterford — the Déisi heartland, with 56 recorded Ogham stones — that sent its people westward. The Pembrokeshire stones are, essentially, the continuation of the Waterford tradition on the other side of the sea.
Two audiences, one stone
The formal structure of the bilingual Welsh stones reveals something subtle about how they functioned. The Latin side and the Ogham side are not straightforward translations of each other. They operate differently.
Gives the name and the patronymic in the traditional Irish formula: MAQI ("son of"), MUCOI ("of the tribe of"). Brief. Genealogical. Asserting Irish identity and lineage. Speaking to those who read Ogham — almost certainly an Irish-speaking community.
Longer. Forms full sentences — "here lies," "of such-and-such people" — using the vocabulary of Roman commemorative epigraphy. Speaking to those who read Latin: the Romanised British population, Christian clergy, anyone with access to Roman administrative culture.
The same person is being presented to two different audiences simultaneously. And the name itself can carry additional information: the Ogham form tends to be more archaic, the Latin-alphabet form reflecting a marginally later phonological state. The bilingual stones are therefore not just bilingual in the sense of two languages — they are sometimes bilingual in the sense of two different moments in the same language's history, frozen side by side on the same stone.
On Charles-Edwards's reading, this is political as much as practical. The Irish settler community in Wales was asserting that their language and their script deserved the same monumental dignity as the language of Rome. To put Ogham next to Latin, on the same stone, in the same format, was to claim equivalence.
Wales also preserves something that Ireland almost never does — inscriptions commemorating women. The corpus is overwhelmingly dominated by male names in the standard MAQI ("son of") formula. Wales is the exception. The stone at Eglwys Cymmin (CIIC 362) reads:
Avitoriges — an Irish name — is the woman herself, the person the stone commemorates; her father Cunignus carries a Brythonic name (Welsh Cynin), a small, telling detail of mixed heritage within a single family. The stone is itself bilingual: alongside the Ogham, a Latin inscription on the same face reads AVITORIA FILI CUNIGNI. It stands today in St Margaret's Church at Eglwys Gymyn — a parish that takes its own name from St Cynin, the same name carried by the father commemorated on the stone within its walls.
In a corpus of hundreds of stones in which women are almost invisible, this inscription is significant beyond its individual case. It suggests that in the Welsh Déisi communities — perhaps because of their settler situation, perhaps because of particular local customs — the social function of these monuments was broader than in Ireland. The stones were not only for men of lineage. Women were worth commemorating too, and Avitoriges is the one name from that world we can still read.
Why were they there?
One of the most intriguing scholarly theories about the Déisi in Wales addresses a question the stones themselves cannot answer directly: why did the Roman or post-Roman British authorities allow an Irish group to found a kingdom on their territory?
The Romans had a long-established policy of settling barbarian warriors — foederati — within Roman territory in exchange for military service. When Magnus Maximus stripped the Roman garrisons from Britain in 388 AD to support his campaign in Europe, the western coasts became vulnerable to Irish raiding. The Irish raids into Wales and western Britain were serious — St Patrick's own account of his capture as a young man in western Britain is one piece of evidence for the scale and violence of this raiding. The re-fortification of coastal positions such as Caernarfon and Cardiff was undertaken specifically in response.
The foederati theory proposes that the Déisi were deliberately settled in Dyfed as a buffer — displaced, militarily capable Irish warriors, invited in to defend the coast against other Irish raiders. The bilingual stones, on this reading, are monuments to men who occupied a formal military-administrative role in the late Roman world, which would explain both the Latin component and the specifically Roman commemorative vocabulary. The soldiers and their families, marking their dead in the format of Rome.
This remains an active area of debate. The stones cannot confirm it. But they are consistent with it — and the coincidence of Irish settlement, Roman-style epigraphy, and a coastline under pressure from Irish raids is at least suggestive.
The distribution of Ogham stones in Wales closely reflects the territory of the Kingdom of Dyfed. More Ogham stones survive in Pembrokeshire than in Wexford — the county directly across the water.
Ancient Connections Project (2023), Ogham Stones: A Welsh-Irish Connection Set in StoneThe Irish exception
Bilingual Ogham-Latin stones are overwhelmingly a British phenomenon. In the whole of Ireland, they are almost entirely absent. The Irish epigraphic tradition kept its scripts separate: Ogham for stone, Latin for vellum. This makes the stone from Killeen Cormac, Colbinstown, Co. Kildare — currently on display in the National Museum of Ireland — one of the most remarkable objects in the entire corpus.
Seven Ogham stones were originally found at the Killeen Cormac site. At least one was broken up and reused in the construction of a surrounding wall in the late 19th century — a fate that befell many Irish monuments in the era of "improving" landlords. The bilingual stone survived. It now holds a prominent place in the Ogham in 3D database as CIIC 19, Colbinstown I, and it remains a stone of active scholarly interest.
The stone itself — a greywacke pillar, 1.54 metres tall — carries an Ogham inscription reading OVANOS AVI IVACATTOS ("of Oän, descendant of Éochad") and, in Roman capitals, a Latin inscription usually read as IVVENE DRVVIDES, possibly "the druids erected this for Jovanis." The two inscriptions are not straightforward translations of one another. Preserved word-endings suggest a date in the first half, or early second half, of the 5th century (McManus, 1991). It is one of the very few examples anywhere in Ireland of this specific bilingual combination — and the only one in vertical Roman capitals.
Its existence poses a question that cannot currently be answered: what was the person commemorated here doing with a connection to the Romano-British bilingual tradition? Were they someone who had spent time in Britain? Someone with family connections to Wales or to the Déisi communities? A cleric trained in a milieu where Latin epigraphy was the norm? The stone does not say. It simply exists — an outlier in the Irish corpus, speaking two languages, offering no explanation.
The Killeen Cormac stone matters precisely because it is exceptional. Its existence confirms that the bilingual tradition was not completely unknown in Ireland — it was simply almost never adopted. Someone in Kildare, at some point in the early medieval period, knew about the Welsh tradition or had exposure to Romano-British commemorative culture, and chose to record their dead in the same way. The rule is confirmed by the exception. Ireland kept its scripts apart; one person, in one field in Kildare, did not. The stone stands in the National Museum, still legible after sixteen centuries, naming a man and his ancestor in two scripts at once — and answering none of the questions about why.
John Francis Shearman documented it in 1873 in the Journal of the Royal Historical and Archaeological Association of Ireland, making it one of the earlier formally recorded Ogham stones in the literature. It has been in scholarly discussion ever since.
See the full find record — museum reference, dimensions, HES data →
What we have and what we've lost
One significantly underappreciated point about the Welsh Ogham corpus concerns the conditions of its survival. Ogham letters carved along the edge of a stone are more vulnerable to erosion than a Latin inscription cut on the face. Some stones that now show no trace of Ogham may originally have had inscriptions that weathered away while the Latin survived. The count we have is almost certainly an underestimate.
Most of the Welsh stones were not found in their original context. The majority were being used as gate posts, fence posts, or building material when they were documented. They were later moved into churchyards — preserved, but displaced. The stones we can study today are survivors, not a representative sample of the original monument tradition. The communities that erected them, the landscapes those communities inhabited, the pathways between Ireland and Wales that their boats travelled — almost all of that is gone. What remains are the stones they left behind, repurposed, re-set, and — eventually — recorded.
The Déisi kingdom of Dyfed did not last forever as an Irish-speaking entity. By the 9th century, the population had largely been absorbed into Welsh-speaking Brittonic culture, the Irish language replaced, the Ogham tradition ended. But for several centuries, a community of Irish origin lived, built, farmed, traded, and commemorated their dead in south-west Wales — and they left the most distinctive body of Ogham inscriptions outside Ireland to show for it.
Wales has the largest concentration of Ogham stones outside Ireland, concentrated in the former Kingdom of Dyfed and closely associated with the Déisi migration from Waterford in the late 4th and early 5th century. The bilingual Ogham-Latin stones are the most distinctive feature of the Welsh corpus and almost entirely absent from Ireland. The one significant Irish exception — Killeen Cormac (CIIC 19, Colbinstown I), Co. Kildare — is now in the National Museum of Ireland and remains a stone of active scholarly interest. The foederati theory for the Déisi settlement is plausible but unproven. Women's names appear in the Welsh corpus in a way that they almost entirely do not in Ireland.
Sources: Charles-Edwards (2000); Ancient Connections Project (2023); Shearman (1873); Ogham in 3D database (CIIC 19, Colbinstown I); McManus (1991); Morus-Baird, Clarke et al., Ancient Connections: The Shared Stories of Pembrokeshire and Wexford; White (2021).
Why it matters
The Welsh stones are the record of a people in transition — between Ireland and Britain, between Ogham and Latin, between an Irish identity and a new one being forged in a different landscape. They chose to preserve both, on the same stone, rather than abandon either. That choice is not just epigraphy. It is a statement about who they were and who they intended to remain.
Ogham in Wales outlasted Ogham in Ireland as a monumental tradition. The script that Ireland largely abandoned for stone by the 7th century continued to be carved in Pembrokeshire because the community that used it had reasons to keep using it — reasons of identity, lineage, memory, and perhaps political legitimacy. A kingdom founded by Irish exiles needed to remember that it was Irish, even as it became Welsh.
The stones in Pembrokeshire's churchyards were not put there to be studied. They were put there to say: someone lived here, of this lineage, of this people. That they say it in two languages is the accident of their particular history. That the history is recoverable at all — from a handful of stones rescued from field boundaries and gate posts — is a small miracle of survival.
Continue reading
Scotland — Dál Riata, the Pictish inscriptions, and the language that has defeated four centuries of scholarship.
Scotland →