Entity
Epona
The Gaulish horse goddess, guardian of horses and their riders, whose cult spread through the Roman cavalry across the western empire and earned her a feast day in Rome itself.
Epona is the Gaulish goddess of horses — patron of the animals themselves and of the people who bred, drove, and rode them. Her name is built on the Gaulish word for horse, epos, carried by the augmentative suffix -on- that turns a noun into a title, so that the name resolves to something like the Great Mare: the horse raised to the order of the divine. Almost everything known of her comes not from any narrative but from the stone and bronze she was left in — roughly two hundred and fifty figurative monuments and more than sixty votive inscriptions, scattered across the lands Rome took from the Celts. She is a goddess one reads from objects. The texts that mention her at all were written by men who did not worship her, and wrote her name to make a point about someone else.
She has no surviving myth, and very likely never had a scripture of any kind. What stands in their place is a picture of remarkable consistency. Two arrangements recur across the whole reach of the cult. In the one favored in the empire’s late shaping, she sits sidesaddle on a walking mare, both turned the same way, composed and unhurried, a rider going nowhere in particular. In the older Gaulish arrangement she stands or sits between horses — often two, sometimes more — set symmetrically to either side, with foals reaching toward the mares. Her hands are rarely empty. She holds a patera, the shallow offering-dish; or a basket of fruit, ears of grain, a sheaf, a loaf; or a cornucopia tipped toward the viewer. Sometimes a foal feeds at the mare’s flank while the goddess looks on; in a few pieces she carries a key, or a folded cloth like the mappa a magistrate dropped to start the races. The horse and the harvest sit together in her iconography, so that the keeper of the stable reads also as a giver of plenty — a pairing the images assert and never explain. Where most gods of the Roman world arrive trailing stories, Epona arrives as a gesture, repeated thousands of times and never narrated.
The two arrangements are not two goddesses. The rider-type and the between-horses type appear side by side in the same regions and the same periods, and the habit of naming one the imperial type and the other the Gaulish type describes a tendency, not a border. What holds across both is the orientation: she faces the animal, or stands among the animals, attending to them rather than to the viewer. She is almost never armed and almost never enthroned in majesty. Even where she sits, she sits as one sits to work near horses, low and close. The dignity in the images is the dignity of competence — the figure who knows the animal and is trusted by it — rather than the dignity of a throne. That, too, is information: the people who commissioned these pieces wanted a goddess who would be at home in a byre, and the sculptors gave them one.
Read from the stones
The absence of myth is not a gap in the record; it is the record. Epona belongs to a religious world that left its asks in stone and bronze rather than in verse, and the modern instinct to supply her a theology runs ahead of what the objects will bear. They are mostly small — household and stable pieces, votive reliefs, statuettes meant for a niche — and they cluster geographically with a sharpness that itself says something. The densest concentrations lie in the horse country of eastern and northeastern Gaul, in what is now Burgundy and the Rhineland, the same ground that bred and traded the animals on which a cavalry empire depended. Where the monuments thin, horses thinned too. The goddess maps onto a working landscape.
The inscriptions sharpen the picture further, because of who set them up. Dedications to Epona come from grooms, muleteers, drivers, ostlers, the keepers and movers of animals, and above all from cavalrymen and the units they served in. They follow the dry, durable formulae of Roman votive practice — the name of the goddess in the dative, the dedicator named, the vow paid willingly and deservedly — the same grammar a soldier would use for any god he owed. Several call her Epona Regina, the queen, and a cluster honor her in the plural, Eponae or Eponabus, as though the one goddess could be many, the way a herd is one thing and many at once. These were not theologians. They asked for the plain things a person asks who lives by horses — that the animals stay sound, that the foals come, that the road be safe and the return certain — and they thanked her in the same plain terms when the asks were met. Read against this, the iconography stops being a puzzle and becomes a job description: the dish and the grain are what the mare needs to thrive, the foal is the increase, the rider is the safe traveler the worshipper hoped to be. The consistency of the image across a thousand miles is the consistency of the work it answered.
Carried by the cavalry
The reach is the surprising part. A deity native to the Gauls became, through the Roman army, a soldiers’ goddess carried wherever her devotees were posted. The vehicle was the auxiliary cavalry — the mounted units raised largely from Gallic and Germanic peoples and then stationed across the frontier. Where those alae went, Epona went: along the Rhine and the Danube, into Britain, across to North Africa, and finally to Rome and into Italy. She was honored in the cavalry’s own quarters; her image had a recognized place in the stable itself, set where the animals were kept rather than in a distant temple. A goddess who began in the byres of Gaul ended garrisoned along the edge of the known world, because the empire that absorbed her ran on the animals she protected.
This is the mechanism behind a fact that can look like mystery. Most Celtic gods stayed home — they were the powers of a particular spring, hill, or people, named for the place and unintelligible away from it. Epona was portable in a way few of them were, and the reason is in her job rather than her theology. Horses moved; the men who tended them moved; and a goddess defined by an animal and a craft rather than by a locality could move with them and still mean the same thing in Britain that she meant in Burgundy. The empire did not so much spread her cult as provide the roads and the postings along which a working devotion already traveled. Where the Roman state took conquered gods, it more often renamed them to fit its own — folding a Gaulish power into a familiar Roman one and worshipping the compound. Epona it largely left as she was, under her own Gaulish name, because what she did had no clean Roman equivalent and needed none.
That diffusion crossed a threshold few foreign gods reached. A surviving rustic calendar — a stone fasti from Guidizzolo, near Mantua in northern Italy — fixes the eighteenth of December as a day for Epona, set within the long midwinter festival of Saturnalia. It is a modest, provincial monument, and the day may have been more local than civic; but it makes her, so far as the calendars record, the one Celtic deity to hold a fixed place in the Roman religious year. The horse goddess of the conquered peoples kept an appointment in the timetable of the conquerors.
The sneer in the record
The literary notices are few, and they share a tone. Roman and Christian writers who name Epona name her to mock — to point at the spectacle of an empire keeping a goddess in its stables. The satirist Juvenal invokes her in passing as the deity sworn by amid the reek of the stalls and the painted images that decorated them, the joke being that a man’s pretensions might be measured against the company his gods kept. Apuleius, in the second-century novel that ends in the rescue of its narrator by Isis, gives the fullest description: his hero, trapped in the body of an ass, comes upon an image of Epona set in a small shrine on the central pillar of a stable, devoutly wreathed with freshly cut roses. The detail is offhand and, for that reason, valuable — it shows the cult as ordinary furniture of a working stable, the kind of thing a passing animal might be expected to find there, kept up with real care. The Christian apologists, among them Tertullian and Minucius Felix, turned the same image to polemic, taunting their pagan neighbors with the worship of a goddess of beasts of burden. None of these writers tells her story. None seems to know one. They confirm only that the image was familiar enough to be a punchline, which means it was everywhere.
The everywhereness is borne out by the material. Beside the inscribed altars and the carved reliefs stand small bronze figurines and clay statuettes, cheap and portable, the devotion of households rather than commissions — the goddess kept in a stable niche, as Apuleius describes, or among the figures of a domestic shrine. The cult had a top and a bottom: an officer’s dedication and a groom’s clay figure could honor the same goddess in the same idiom. That spread across classes, and not only across provinces, is why so much of the evidence is modest. Epona was not principally a goddess of great temples. She was a goddess of the places where horses actually were, and the surviving objects keep the shape of that fact.
What the silence will and will not carry
The plainness of the stones has not stopped fuller reconstructions, and the two most persistent are worth naming for what they each rest on. One reads Epona as a sovereignty goddess — the divine partner whose bestowal of the horse, or whose union figured as a mare, confers legitimate rule, drawing the line forward to the horse-women of later Welsh and Irish story. The other reads her as a psychopomp, a guide of souls, taking the riding figure for a conductor of the dead toward the Otherworld and pointing to horses in funerary art. Both have distinguished advocates; both reach well past the evidence. The sovereignty reading borrows its theology from texts written centuries later in another country and projects it backward onto silent reliefs. The psychopomp reading leans on a handful of grave contexts and on the general truth that the dead, in many places, travel — neither of which singles Epona out. The cautious account, and the one most who work directly with the corpus hold, stays level with the monuments: a protector of horses and of the people whose lives moved on hooves, asked to keep the animals sound and the road safe, and thanked when she did.
The harvest attributes she carries fit that account without strain. A goddess of the stable is a goddess of an economy, and in an agrarian world the line from the sound mare to the full granary is short and literal: the same beasts plow, haul, and breed the wealth the cornucopia stands for. The grain in her hands is not a clue to a hidden fertility cult. It is the visible end of what the horse is for.
The shape of the evidence
The serious study of Epona is a study of objects, and it begins where the objects were first gathered and counted. Salomon Reinach assembled and scrutinized the early corpus in his essay Épona, published in the Revue archéologique in 1895, weighing which images could safely be called hers and which could not — the founding act of caution that the field has, on its best days, kept. The monuments themselves were cataloged within the vast inventory of Gallo-Roman relief sculpture compiled by Émile Espérandieu in his Recueil général des bas-reliefs, statues et bustes de la Gaule romaine, where Epona’s images take their place among the thousands of stones that survive the religion of Roman Gaul. The dedicated catalog is the 1953 Epona: déesse gauloise des chevaux, protectrice des cavaliers of René Magnen and Émile Thévenot, which fixed the figure-types and listed some two hundred and fifty-three monuments; the definitive modern survey is Marion Euskirchen’s exhaustive corpus and study, Epona, in the Bericht der Römisch-Germanischen Kommission 74 (1993), which re-counted, re-dated, and re-described the whole body of evidence with the find-context of each piece recorded. The contested theologies have their literature too: the funerary-guide reading is associated with mid-century French scholarship, and the identification of Epona with Isis that Apuleius seems to invite is examined by Jeffrey Winkle in Epona Salvatrix?, which sets the novel’s stable-shrine scene against the iconographic record. Each of these works ends, in the end, at the same place: with images that show much and say nothing.
That is the discipline the subject imposes. Epona is the rare deity who comes down almost entirely as picture and almost not at all as word — a face turned the same way across a thousand reliefs, a hand always full of the same dish or grain, a mare always walking. The temptation is to fill the silence with a story worthy of the spread, to make the Great Mare carry kingship or carry the dead. The stones decline the offer. They hold to the gesture: the woman seated among horses, feeding them, going with them, asked for nothing grander than that the animals live and the rider come home — and answering, as far as the people who left her can tell us, in exactly those terms.
→ Related: Pomona · Bona Dea · Apuleius · Gaul · Roman Empire
Sources
- Reinach 1895 (Revue archéologique)
- Espérandieu, Recueil général
- Magnen & Thévenot 1953
- Euskirchen 1993 (BRGK 74)
- Winkle 2015 (Ancient Narrative)