Our Approach · Origins · Western Europe

Gaul, Britain, Ireland, and Wales · c. 420 BCE to 1405 CE

Celtic and Druidic

Irish law fixed who must feed, house and quiet an injured person, and for how long.

Celtic healing survives in four kinds of evidence: Roman reports written by conquerors, votive deposits at sacred springs, early Irish law tracts, and medieval Welsh recipe books. The law carries the most weight. Bretha Crólige gave an injured person a nursed, quiet, financed recovery at the offender's expense. Irish jurists were regulating the conditions under which a damaged body restores itself. The Unified Model of Tone names the state they were protecting tone.

Celticart
forthcoming

Documented span

c. 420 BCE to c. 1405 CE

Region

Gaul, Britain, Ireland, and Wales

Key text

Bretha Crólige, edited by D. A. Binchy in 1938

Oldest hard evidence

Trepanned skulls, Münsingen-Rain, 420 to 240 BCE

The documentary record

What the record shows

Almost everything popularly repeated about druid medicine traces to two unreliable places: Greco-Roman writers describing an enemy, and Irish and Welsh scribes writing centuries after conversion. These are the dated items that survive scrutiny.

Julius Caesar, campaigning in Gaul in the 50s BCE, gives the fullest early account of the druids in Book 6 of the Gallic War. He reports that they settle almost all disputes, and that trainees "learn by heart a great number of verses" for as long as twenty years. (McDevitte translation)

The primary witness for the druids is the general conquering them, and his account never calls them healers.

Pliny the Elder, writing about 77 CE, records that the principate of Tiberius suppressed the druids of Gaul, whom he classes as "hoc genus vatum medicorumque", this class of seers and physicians. (Natural History 30.13)

The textual basis for calling druids physicians is one clause inside a Roman attack on magic.

Pliny describes the mistletoe rite at Natural History 16.95: the sixth day of the moon, two white bulls, a priest in white, a golden sickle. In Bostock and Riley's 1855 translation the name meaning "the all-healing" attaches to the moon rather than to the plant. (Bostock and Riley, 1855)

The most quoted fact about druid medicine rests on an ambiguous phrase in a second-hand Roman report.

Two skulls from the La Tène cemetery at Münsingen-Rain in Switzerland carry surgical openings, and they are dated 420 to 240 BCE. Across all periods 34 trepanned individuals have been published from Switzerland, and in the Late Iron Age group the survival rate is 78 percent. (Moghaddam et al., 2015)

Iron Age Europe opened skulls and the patients healed, which is surgery with a technique behind it.

A burial at Stanway outside Camulodunum was dug between about 40 and 50 CE and excavated between 1987 and 2003. It held copper alloy surgical instruments, a strainer bowl carrying traces of Artemisia, and a set of metal divining rods. (Colchester Archaeological Trust)

The one British grave with a medical kit also carries divination gear, so healer and diviner were the same office.

Excavations of the sacred spring at Bath in 1979 and 1980 recovered about 130 inscribed lead and pewter tablets of the second to late fourth century CE. They entered the UNESCO Memory of the World UK Register in 2014. (Roman Baths, Bath)

The largest body of writing from a Celtic sacred spring is about stolen property rather than about illness.

Bretha Crólige, the Judgments of Blood-Lying, belongs to the Senchas Már and dates to the late seventh or early eighth century CE. It survives in one manuscript, National Library of Ireland MS G 11, and was edited and translated by D. A. Binchy in Ériu 12 in 1938. (Kelly, 2001)

The strongest evidence for Celtic care is a statute, and the whole tract hangs on a single surviving copy.

The Welsh medical recipe collections tied to Myddfai survive in four manuscripts of the late fourteenth or early fifteenth century. The Red Book of Hergest is among them, copied between 1382 and 1405. (Luft, 2020)

The Welsh material postdates the druids by more than a thousand years and belongs to European learned medicine.

The source problem

Why almost nothing about druid medicine can be verified

Four bodies of evidence carry the Celtic healing record, and each is compromised in a different way. Roman authors wrote about people Rome was conquering or had already suppressed. Sacred springs preserve objects and prayers, not procedures. Irish sagas reach us through Christian scribes writing centuries later. The Welsh recipe books are fourteenth century at the earliest.

The druids themselves are silent by choice. Caesar reports that they refused to commit their teaching to writing while the Gauls around them used Greek letters for ordinary business. When the last trained memory died, the content went with it.

That leaves law and archaeology carrying the weight, and both are stronger than the material usually quoted. A statute cannot flatter its subject. A trepanned skull with healed margins cannot be a later legend.

Caesar, 50s BCE

What Caesar actually reports about the druids

Caesar places the druids in Book 6 of the Gallic War as one of two ranks that matter in Gaul. They conduct public and private sacrifices, interpret religious questions, and decide almost all disputes, from homicide to inheritance to boundaries.

The training is the detail that stuck. Students learn a great number of verses by heart, and some remain twenty years under instruction. Caesar also records the teaching that souls pass from one body to another after death, along with lessons on the stars and the nature of things.

What Caesar never says is that druids treated the sick. Priest, judge, teacher, and astronomer appear in his list. Physician does not. Every account of the druid as healer is inference.

Pliny, c. 77 CE

The mistletoe rite and the phrase everyone quotes

Pliny the Elder gives the one Celtic healing rite that can still be read in detail. At Natural History 16.95 the Gauls gather mistletoe from the oak on the sixth day of the moon. A banquet is laid beneath the tree and two white bulls are brought, their horns bound for the first time. A priest in a white robe climbs and cuts the plant with a golden sickle. Pliny adds that the mistletoe was held, taken in drink, to grant fertility to barren animals and to counter every poison.

The famous phrase is less secure than its fame. Pliny writes omnia sanantem appellant suo vocabulo, they call it in their own language the all-healing. The Latin participle is masculine or feminine in form, so it cannot agree with the neuter word for mistletoe, and Bostock and Riley's 1855 translation therefore attaches the name to the moon.

Pliny names two other druidic plants at Natural History 24.103 and 24.104. Selago is picked without iron by a person in white with bare washed feet, and its smoke treats the eyes. Samolus is gathered left-handed by someone fasting, and it treats disease in pigs and cattle. Both entries specify procedure. Neither gives a dose.

The Druids, for that is the name they give to their magicians, held nothing more sacred than the mistletoe and the tree that bears it.

Pliny, Natural History 16.95 · translated by John Bostock and H. T. Riley, 1855

Aquae Sulis

What the sacred spring at Bath was actually used for

Hot mineral water rises at Bath, and the presiding deity there was Sulis. Rome recognized its own Minerva in her and worshipped Sulis Minerva at Aquae Sulis, building a temple and a bathing complex around the spring. The monumental architecture is Roman work laid over a British cult site.

Excavations of the spring in 1979 and 1980 recovered about 130 inscribed tablets of lead and pewter, dating between the second and the late fourth century CE. Scholars call them prayers for justice. The great majority concern theft, with the goddess asked to act against whoever holds a stolen cloak, ring, or purse.

In these texts health works as the lever rather than the complaint. Several tablets ask that the thief be denied sleep, health, or rest until the property is returned. The deposits record a belief that bodily state could be withheld and given back, and they record no treatment at all.

The source of the Seine

Sequana's sanctuary and the shapes of the sick

The sanctuary of the goddess Sequana stood thirty kilometers north of Dijon, where the Seine rises. The complex runs about 100 meters and holds a spring, a piped catchment, a basin inside a peristyle courtyard, and a small fanum of Celtic type. Most of the buildings date no earlier than the second century CE, and coins show the site active into the fourth.

Its votive sculpture is the richest collection in Gaul. Excavation began in 1836, and the campaigns through 1967 produced roughly 1,500 objects in stone, bronze, and wood. Several hundred wooden pieces came from the marshy ground in the seasons of 1963, 1966, and 1967 alone, preserved by waterlogged soil. A bronze Sequana standing in a duck-prowed boat was found in 1933 and is kept at the archaeological museum in Dijon.

The offerings prove a healing cult because they are anatomical. Limbs, torsos, heads, and internal organs appear, with bronze plaques of eyes and breasts and a series of swaddled infants. A pilgrim carried the shape of the affected part to the water and left it. The gesture names the part precisely, then hands over the whole person.

Bretha Crólige, c. 700 CE

Early Irish law made convalescence a debt

Ireland was never conquered by Rome, and its lawyers wrote things down. Bretha Crólige, the Judgments of Blood-Lying, sits inside the Senchas Már and dates to the late seventh or early eighth century. One manuscript preserves it, and D. A. Binchy edited and translated it in Ériu 12 in 1938.

The tract governs othrus, sick-maintenance. Nine days after an unlawful injury the wound is formally assessed, and four outcomes follow. The victim dies and the culprit owes the homicide fine. The victim recovers and is compensated by the class of wound. The injury is judged mortal and crólige báis falls due. Or recovery is expected, and the victim must be taken away and nursed.

In that fourth case the culprit carries the whole cost. He removes the injured person to a house that is not his own and pays for the nursing. He feeds the attendants, supplies a substitute worker, and entertains the visitors who come every ninth day. Recovery is treated as a thing with duration, conditions, and a price, and the law protects all three.

There are not admitted to him into the house fools or lunatics or senseless people or half-wits or enemies. No games are played in the house. No tidings are announced. No children are chastised.

Bretha Crólige · translated by D. A. Binchy, Ériu 12, 1938

The sick-house

The statute regulates the room, not only the payment

Bretha Crólige places the injured person in a house with four doors so that he can be seen from every side, with water running across the middle. Fools and quarrelsome company are barred, and so is food the physician has forbidden.

The prohibitions run through noise and disturbance one item at a time. No games. No news announced. No children beaten. No blows exchanged between men or women. No dogs set fighting nearby. No shouting, no squealing pigs, no brawls, no screaming.

Early Irish law prices quiet the way it prices cattle. A companion tract, Bretha Déin Chécht, governs the physician instead of the patient. It sets the fee at half the injury payment for wounds to the twelve doors of the soul.

Cath Maige Tuired

What the sagas remember, and how they remember it

Irish myth kept the healer at the center of sovereignty. In Cath Maige Tuired, the Second Battle of Mag Tuired, King Núadu loses a hand and cannot rule while blemished. The physician Dian Cécht fits him a silver hand that moves like any other. Dian Cécht's son Míach then grows the true hand back over nine days and nights: three days against the side, three against the chest, three under wisps of burnt bulrush.

The same text holds the well called Sláine. Dian Cécht, his sons Octriuil and Míach, and his daughter Airmed chant over it, and the mortally wounded thrown in come out alive. After Míach's death three hundred and sixty-five herbs grow from his grave, one for each of his joints and sinews. Airmed sorts them on her cloak and her father scatters them so that nobody knows their properties.

These are sagas, not case notes. Elizabeth A. Gray's translation works from British Library Harleian MS 5280, centuries later than the events described. What the text preserves is the shape early Ireland gave to healing: staged, counted in nine-day units, ranked among the highest arts.

Meddygon Myddfai

What the Welsh recipe books actually contain

Welsh tradition names a family: Rhiwallon and his sons of Myddfai in Carmarthenshire, said to have served Rhys Gryg of Dinefwr, who ruled between 1195 and 1234. A preface in the Red Book of Hergest ascribes a collection of remedies to them. John Pughe's English translation of 1861 made the Physicians of Myddfai famous.

Diana Luft's 2020 edition for the University of Wales Press dismantles the tidy version. There is no single text called Meddygon Myddfai. Four related manuscripts of the late fourteenth or early fifteenth century each carry their own collections, and the Myddfai preface covers only the collection that immediately follows it.

The contents are ordinary European learned medicine written in Welsh. Bloodletting calendars. Lists of dangerous days. A prognosis taken by binding bruised violet to the legs. A regimen chapter warns that whoever eats, drinks, sleeps, or labors more or less than he should will not escape sickness. Luft traces strong links to contemporary English and continental medical writing.

After the last druid

Modern Druidry begins in the eighteenth century

The druids were suppressed under the emperors, Tiberius among them by Pliny's account, and their teaching ended with the last trained voice. What continued was everything they had never controlled: holy wells, Irish vellum, Irish law, Welsh recipe books.

Ronald Hutton's Blood and Mistletoe, published by Yale University Press in 2009, traces how the gap was filled. William Stukeley attached the druids to Stonehenge, a monument raised roughly two thousand years before them. Iolo Morganwg forged supposedly ancient bardic texts and invented the Gorsedd ceremonies that entered the Eisteddfod.

Modern Druidry is a living religion with its own integrity and its own eighteenth-century origin. It is not a survival, and it is not evidence about the Iron Age. Most popular druidic healing lore descends from the revival.

The convergence

What the Celtic record documents, and what it is called now

Celtic sources are thin on theory and unusually strong on procedure. The Irish law tracts never explain what healing is. They specify what must be arranged around a damaged body, for how long, and at whose expense. Independent traditions across the world converged on the same short list of observations, which is evidence that the thing being observed was real and readable without instruments. The Unified Model of Tone names that state tone and measures it.

  • Bretha Crólige sets a nine-day assessment, then four prognoses, then a nursing period that runs until the injury resolves.Time course, the shape recovery takes in time, which any treatment has to respect.
  • Sick-maintenance moved the entire burden onto the offender: the nursing, the food, the attendants, and a substitute to do the injured person's work.Load, the accumulated demand a body carries while it is trying to restore itself.
  • The sick-house is defined by what reaches the patient. Running water across the floor, four doors, no shouting, no fighting dogs, no forbidden food.Input quality, where the character of an input decides the response it produces.

Common questions

Frequently asked

Did the druids leave any writings of their own?

No. Caesar reports in Book 6 of the Gallic War that the druids refused to commit their teaching to writing, while the Gauls around them used Greek letters for ordinary business. Trainees learned verses by heart and some stayed twenty years in training. Everything about them therefore comes from Greek and Roman authors, from archaeology, or from Irish and Welsh manuscripts copied centuries later by Christian scribes. Each of those sources distorts in a different direction. The gap between the druids and every witness to them is the whole problem.

Were the druids physicians?

The evidence is one clause. Pliny records at Natural History 30.13 that the principate of Tiberius suppressed the druids of Gaul, whom he groups as seers and physicians, and elsewhere he credits them with the plants selago and samolus. Caesar, the fullest early witness, describes them as priests, judges, teachers, and astronomers, and never as healers. Healing sat inside a learned class that also handled law and ritual, which is as far as the sources go. Everything past that point is inference.

What is Bretha Crólige and what did it require?

Bretha Crólige is an early Irish law tract of the late seventh or early eighth century, part of the Senchas Már, edited by D. A. Binchy in Ériu 12 in 1938. It governs othrus, sick-maintenance. Nine days after an unlawful injury the wound is assessed. If recovery is expected, the culprit removes the victim to a neutral household and pays for nursing, food, attendants, and a substitute worker. The tract also specifies the sick-house itself, down to the noise allowed inside it.

Was mistletoe really an all-healing medicine?

Pliny reports a belief, and the report is shakier than its fame. At Natural History 16.95 the plant is cut from the oak with a golden sickle on the sixth day of the moon. It was held to grant fertility to barren animals and to counter all poisons. The Latin phrase meaning the all-healing is grammatically masculine or feminine, so Bostock and Riley's 1855 translation attaches it to the moon. Mistletoe is toxic in quantity, and the passage is testimony about belief rather than pharmacology.

What are the Physicians of Myddfai?

A Welsh family of physicians said to have served Rhys Gryg of Dinefwr, who ruled from 1195 to 1234, and the name attached to a body of medical recipes in Welsh. Diana Luft's 2020 edition shows there is no single text called Meddygon Myddfai. Four manuscripts of the late fourteenth or early fifteenth century each carry their own collections, and the Myddfai preface covers only the collection following it. The contents track contemporary English and continental medicine, and the name is later than the recipes.

Is modern Druidry a continuation of the ancient druids?

No. Ronald Hutton's Blood and Mistletoe, published by Yale University Press in 2009, traces modern Druidry to an eighteenth-century revival with no line of descent from the Iron Age. William Stukeley attached the druids to Stonehenge, a monument raised about two thousand years before them. Iolo Morganwg forged bardic texts and invented the Gorsedd ceremonies. Modern Druidry is a living religion of its own, and it is not a source of information about ancient practice. Its origin is dated and documented.

What does the Unified Model of Tone say about Celtic healing?

That the Irish jurists were legislating for the conditions of recovery, and that those conditions are measurable now. A nine-day assessment followed by a nursing period is a claim about time course. Transferring the work, the food, and the cost to the offender removes load. Specifying quiet, company, water, and diet inside the sick-house is a claim about input quality. The model does not adopt any Celtic explanation. It names and measures the state the statute protected. The Irish observations stand on their own.

Keep reading

Where this connects

  • Tone is the regulated state the Irish sick-maintenance law was built to protect, defined and measured.
  • Time course is why a nine-day assessment and a fixed nursing period were the right shape for an injury.
  • Load explains why removing a person's work and costs mattered as much as dressing the wound.
  • Input quality is why the statute regulated noise, company, water, and food inside the sick-house.
  • Ancient Greece ran healing sanctuaries at springs on similar logic, with far better records.
  • Norse and Germanic is the neighboring tradition whose surviving medical books specify preparation, dose, and duration.
  • Slavic and Baltic shares the same evidence problem, a record written by outsiders and later scribes.

Sources

References

  1. Binchy, D. A. Bretha Crólige. Ériu 12 (1938), 1 to 77. Edition, translation, and commentary. Listed in the Ériu contents index.
  2. Binchy, D. A. Sick-maintenance in Irish law. Ériu 12 (1938), 78 to 134.
  3. Kelly, F. Medicine and early Irish law. Irish Journal of Medical Science 170(1), 2001, 73 to 76. PMID 11440420.
  4. Grace, P. A. Subtracting insult from injury: the medical judgements of the Brehon law. History Ireland.
  5. Caesar. De Bello Gallico, Book 6, chapters 13 and 14. Translated by W. A. McDevitte. Project Gutenberg.
  6. Pliny the Elder. The Natural History, Book 16, chapter 95. Translated by John Bostock and H. T. Riley. Taylor and Francis, 1855.
  7. Pliny the Elder. Naturalis Historia, Books 24 and 30, Latin text. LacusCurtius, University of Chicago.
  8. Gray, E. A., translator. Cath Maige Tuired: The Second Battle of Mag Tuired. CELT, University College Cork.
  9. Luft, D. Medieval Welsh Medical Texts, Volume One: The Recipes. University of Wales Press, 2020.
  10. Pughe, J., translator, and Williams ab Ithel, J., editor. The Physicians of Myddvai. Welsh Manuscripts Society, 1861.
  11. Moghaddam, N. et al. Survival after trepanation: early cranial surgery from Late Iron Age Switzerland. International Journal of Paleopathology 11 (2015), 56 to 65.
  12. Colchester Archaeological Trust. Druids at Colchester: the Stanway burials. See also Crummy, P. et al. Stanway: An Elite Burial Site at Camulodunum. Britannia Monograph 24, 2007.
  13. The Roman Baths, Bath. Roman curse tablets from the sacred spring.
  14. UNESCO. Bath's Roman curse tablets added to the UNESCO Memory of the World.
  15. Rolley, C. Sequana (Saint-Germain-Sources-Seine). The Princeton Encyclopedia of Classical Sites. Princeton University Press, 1976.
  16. Hutton, R. Blood and Mistletoe: The History of the Druids in Britain. Yale University Press, 2009.

Last updated 2026-08-10