In Mayo, the saints do not only bless. They also curse. Several of the parishes and families of north Connacht carry traditions of saintly displeasure — curses laid down in the early Christian centuries and passed down through the oral record as an explanation for later misfortune. The O’Dubhda have one such tradition, attached to St Gerald of Mayo.

This page sets out what we know about Gerald, what the clan tradition says about the curse, and where the history and the folklore diverge. This is a working page. If you carry a family version of the curse story, or know of a printed source that documents it, we want to hear from you.

I.  Who Gerald was

St Gerald (d. 732) is a historically attested figure. He was an Anglo-Saxon — Northumbrian by birth — and came to Ireland with St Colman of Lindisfarne after the Synod of Whitby in 664, when the English church sided with Rome and Colman, who held for the older Irish and Columban traditions, withdrew with his monks to Iona and then to Mayo.

Colman first settled the Irish party on Inishbofin, off the west coast of Mayo. A quarrel over labour arose between the Irish and English monks, and Colman then founded a separate settlement for the English monks on the mainland. This house — Mayo of the Saxons, Magh Eo na Sacsan — is what gave the county its name. Gerald became its second or third abbot and is traditionally regarded as the foundation’s greatest figure. Bede mentions the settlement in his Ecclesiastical History, as an English community transplanted into Irish soil.

Gerald’s dates place him two to three centuries before the O’Dubhda emerge as a distinct sept. The Uí Fhiachrach Muaidhe — the kindred from which the O’Dubhda later crystallised — were already settled in Tirawley and Tireragh in his day, but the family name Ó Dubhda (“descendant of Dubhda”) does not appear in the annals until the tenth century.

II.  What the tradition says

The form in which the curse survives in family telling is brief: that St Gerald cursed the O’Dubhda, and that the long slow loss of clan power from the late sixteenth century onwards — the Burke takeover, the forfeitures, the dispersal — was the working-out of that curse. Marion Dowd’s October 2025 address to the clan named St Gerald’s Curse as one of the recurring folk explanations for the O’Dubhda collapse, alongside the mermaid’s parting curse at Scurmore.

What the tradition does not agree on is the grievance. Variants heard in family tellings include:

• That an O’Dubhda chief plundered the lands of Mayo Abbey in the medieval period and drew the curse on his line.
• That an early O’Dubhda forebear refused Gerald hospitality, or refused to return property to his monks.
• That the curse was not a personal act by Gerald but the malediction of the saint’s community — the monks invoking his name against a later injury.

Each of these is typical of the “saint’s curse” pattern across Gaelic Ireland. The Martyrology of Donegal and the Book of Lismore both record saints’ curses of this shape: a prince dishonours a monastery, the abbot delivers a formal malediction, the prince’s line fails. The pattern is so regular that it becomes a grammar for explaining historical loss.

III.  What the record shows, and does not

Not in the annals. The Annals of the Four Masters, the Annals of Connacht, and the Leabhar Mór Leacain (the Great Book of Lecan, compiled by the Mac Fhirbhisigh scholars of Lackan) between them provide a more or less continuous record of the O’Dubhda chiefs from the tenth century onwards. None of those sources preserves a direct account of St Gerald cursing the family. The O’Dubhda are well documented as patrons of ecclesiastical foundations — Moyne, Rosserk, Killala, and Ardnaree all received clan support at different points — but there is no reference in the surviving annalistic record to a rupture with Mayo Abbey of the sort the tradition describes.

Not in O’Donovan 1844. John O’Donovan’s 1844 edition of the Genealogies, Tribes and Customs of Hí Fiachrach, compiled from the Mac Fhirbhisigh manuscripts, is our most complete published source for the clan’s history through 1664. It does not mention St Gerald’s curse. The seventeenth-century manuscript of The Book of the O’Dowdas (Mac Fhirbhisigh), transcribed into the 1844 edition, likewise does not.

A thin surviving trace. This is not the same as saying the tradition is fabricated. Curses of this kind typically survive in oral tradition and in parish lore, not in the annals or the genealogical tracts. The Schools’ Collection of 1937–1939 may hold an entry that names both St Gerald and the O’Dubhda together; we have not yet located one. Mac Hale’s 1990 survey of north Connacht sites may mention it; the relevant volume has not yet been systematically mined for folk material.

IV.  How to read the tradition

Whatever the source, the tradition is doing real cultural work. In the three centuries between the Burke takeover of the early 1600s and the Schools’ Collection of the 1930s, the Connacht Irish needed to explain their own dispossession. Several explanations were in active circulation. The mermaid left because the cloak had been stolen; the land had been lost as a dancing-shoe wager in Dublin; a saint, long dead, had cursed the line. All three are different registers of the same underlying question: how did a kingdom fall?

To read St Gerald’s Curse naively — as a verifiable act by an eighth-century abbot against a family that did not yet exist by name — is to miss the point. To read it as a community’s way of processing its own loss, recruiting a local saint to the work of explanation, is to read it at its depth.

V.  Sources flagged

Published sources on Gerald that document his historicity but not the curse:

• Bede, Historia Ecclesiastica Gentis Anglorum (c. 731) — records Colman’s withdrawal to Mayo.
• Martyrology of Donegal (compiled 1630) — 13 March, feast of “Gerald of Mayo.”
• J. M. Flood, Ireland, its Saints and Scholars (1917) — summary life, no curse.

A standing request: if you have ever heard a specific version of St Gerald’s Curse — or know of a printed source that gives it — please send it along. This page will grow as the evidence does.