Cernunnos: The Celtic Horned God Who Is Named Only Once

Cernunnos is a Gaulish god named on one surviving monument: a stone pillar raised in Roman Paris between 14 and 37 CE. The stone now preserves only ERNVNNOS, although eighteenth-century drawings show a C, so scholars cannot tell whether the first antiquarians supplied it or the letter was lost later. Every other antlered figure linked to Cernunnos, including the famous one on the Gundestrup Cauldron, carries no name at all.

What is the actual evidence for Cernunnos?

It fits on a short table. Antlered figures are not rare in Celtic and Gallo-Roman art, and similar-looking divine or personal names appear elsewhere. What is unique is an antlered figure paired with this name, and even that label is damaged.

Only the first row combines an antlered image with this name, and its inscription is incomplete today. Everything else is either a picture with no name or a similar-looking name with no secure connection to the image. The god most people can describe from memory rests on a grouping scholars made from resemblance. That grouping is not arbitrary, but it cannot settle whether these images show one god or several related ones known by similar names.

Where does the name Cernunnos come from?

From a single monument. The Pillar of the Boatmen was raised in Lutetia, Roman Paris, during the reign of Tiberius, by the local guild of river boatmen. It was rediscovered in 1711 beneath the choir of Notre-Dame de Paris, and it now stands in the Musée de Cluny. It is a genuinely strange object: Roman gods and Gaulish gods share its faces, with Jupiter, Vulcan, and Castor and Pollux on some panels and native deities such as Esus and Smertrios on others.

On one of the lower faces there is an antlered head, and above it a short word in Gaulish. That word is the entire textual basis for the god.

And it is damaged now. Modern examination reads only [.]ERNVNNOS, but the history of the missing initial is not settled. Both reports from 1711 draw a clear C and neither marks it as a conjecture. Alexandre Léonet's 2022 study lays out three possibilities: the first antiquarians supplied the letter, the C was visible in 1711 and disappeared through later damage, or the uneven spacing means that one, two, or even no letters are missing. The firm conclusion is narrower than the usual story: the stone today cannot verify the initial. Later identifications of other antlered figures still depend on stylistic resemblance rather than another labeled image.

What does the name Cernunnos mean?

The usual interpretation is roughly the horned one. The proposed Gaulish root is related to words for horn, including Latin cornu and Greek keras. A Montagnac inscription mentions Carnonos, and a bronze plaque from Steinsel names Cerunincus.

Those spellings are linguistic neighbors, not extra captions. Carnonos may be a personal name, and Cerunincus appears to be a local god. Léonet cautions that nothing proves either one belongs to the antlered figure from Paris. Etymology can suggest what a name meant, but it cannot merge separate cults or images by itself.

Is the Gundestrup Cauldron Cernunnos?

The single most reproduced image of Cernunnos is not labeled Cernunnos by the museum that owns it.

The Gundestrup Cauldron is a large silver vessel pulled out of the bog Rævemosen, in Himmerland, Jutland, in 1891. One of its inner plates shows a cross-legged figure with antlers, holding a torc in one hand and a ram-horned serpent in the other, surrounded by animals. It is the picture in almost every article about the Celtic horned god, including, quite often, articles that never mention that it was found in Denmark.

The National Museum of Denmark, which holds it, is careful. Its description says the cauldron came from a distant area to the south or southeast, that exactly where it was made remains an open question, and that its plates show elephants, lions, and several unknown gods in a foreign style. The museum does not identify this plate as Cernunnos.

The identification is an argument from resemblance: antlers and torcs link the cauldron figure to the Paris relief, while the cross-legged posture resembles other figures placed in the same scholarly family. That posture is not preserved on the Paris stone, because the lower half of its panel is missing. This is a reasonable scholarly comparison, but it is not an ancient caption.

What was Cernunnos the god of?

No surviving ancient text explicitly defines his functions.

The pillar gives an image and a damaged name. It gives no epithet, no function, no myth, no offering, no festival. No surviving Gaulish narrative text fills the gap. Everything said about what Cernunnos governed has been supplied from outside this evidence, and it has been supplied repeatedly, in different directions, by serious people.

Léonet traces the sequence of scholarly identifications: Bacchus first, then Pan, then a god of hunting, then Dis Pater and the underworld. Phyllis Fray Bober's 1951 study in the American Journal of Archaeology, still the foundational treatment of the material, read the antlered god across the Gallo-Roman evidence as a deity of fertility and abundance. Each reading is defensible. None of them is attested.

On the Reims relief, an antlered figure sits between Apollo and Mercury, with a bag spilling what may be grain or coins between a stag and a bull. Ceisiwr Serith proposes that this kind of paired imagery presents Cernunnos as a mediator between wild and domestic life, with a role in wealth and exchange. This is an interpretation of the images, not a function explicitly recorded in an ancient text.

The familiar modern description, lord of the animals, is an accurate summary of what the pictures contain. It is not a title anyone in antiquity is recorded as giving him.

Is Cernunnos related to other horned gods?

Relatives keep getting proposed for him. The list runs to a Bronze Age stag-god from Anatolia, a seal from the Indus Valley, and a ghost in Windsor Forest. Every one of those is a resemblance somebody noticed. No ancient text links Cernunnos to any of them, and most are separated from him by a thousand years or a few thousand miles.

The Anatolian comparison draws on both language and imagery. In his 2014 study, Fred Woudhuizen argued that the Luwian stag-god Kuruntas and Cernunnos share an Indo-European root meaning horn and represent related forms of the same god. He cites earlier linguistic work by Calvert Watkins. Maxim Fomin's 2016 study also compares Cernunnos with Anatolian stag-deity images and inscriptions. These studies propose a shared Indo-European background; they do not establish that ancient worshippers treated the two gods as interchangeable.

The Indus seal shows how a proposed identification can become a familiar label. Marshall read the seated horned figure on seal 420 as an early form of Shiva in his aspect as Pashupati, lord of beasts, and the reading hardened into a caption that still gets printed as fact. Doris Srinivasan tested it in Archives of Asian Art in 1975/1976 by going back to the Vedic texts where the word pashupati lives. A pashupati there guards domestic animals on the road. The creatures around the seal figure are an elephant, a tiger, a rhinoceros, and a buffalo, all of them wild. She concluded that Marshall's identification needs revision, and proposed a divine bull-man instead.

That is the label the previous section ended on. It describes a picture accurately and then gets used as though it were a source. The same phrase landed on a Gaulish relief and on an Indus seal because both show a figure with animals around it. Herne is a third version of the move. Murray put him in the Cernunnos family in 1931, but the earliest written account of Herne is Shakespeare's own, in The Merry Wives of Windsor around 1597, where he is the ghost of a Windsor forest keeper who walks at midnight wearing ragged horns.

The surviving monuments also show which gods appeared beside the figures identified as Cernunnos. The Paris pillar places the named antlered figure on the same monument as Esus and Smertrios, alongside Jupiter, Vulcan, and Castor and Pollux. At Reims, an unnamed antlered figure sits between Apollo and Mercury. Paris preserves the name; the identification at Reims depends on the imagery. Sharing a monument does not establish a family relationship or a particular myth.

None of this shows the antlered figures are unrelated. Horned gods are genuinely common across the ancient world, and Indo-European comparison is a serious method with real results behind it. It just cannot supply the thing missing here, which is a text.

Is there an Irish Cernunnos?

No Irish figure can be securely identified as Cernunnos from the surviving evidence. His name belongs to the Gaulish material; the Irish narratives were written down centuries later. Comparisons have been proposed for more than a hundred years.

Henri d'Arbois de Jubainville proposed that the Ulster heroes Cúchulainn and Conall Cernach were Irish versions of two figures on the Paris pillar, Smertrios and Cernunnos. The proposal depended partly on linking Conall's epithet Cernach with the word for horn. John Arnott MacCulloch rejected that argument in 1911. He noted that Conall was not described as horned and discussed an explanation of Cernach through the angular shape of his head after a blow, as well as the possible meaning victorious. His verdict on this identification was that it is "more ingenious than convincing."

The Dagda offers another possible comparison through the name Fer Benn. The dictionary entry for benn includes the horn of an animal, but also a peak or a point. In Cath Maige Tuired, a woman demands that the Dagda carry her on his back, and he insists that she first call him by his name. He gives Fer Benn, then denies that it is his name and extends it to Fer Benn Brúach before giving a longer sequence. The exchange does not settle which sense of benn is intended, and it does not identify the Dagda as Cernunnos.

The closest Irish parallel to the Gundestrup picture has no horns in it at all. In "Finn and the Man in the Tree", edited by Kuno Meyer in 1904 from a Trinity College manuscript, Finn finds Derg Corra sitting in a treetop with a blackbird on his shoulder and a bronze vessel in his hand holding a live trout, a stag standing at the foot of the tree, and he shares nuts, apple and water with all three. The picture is a man among animals that eat alongside him. The story around it is about a servant hiding from his lord, and nobody in it is called a god.

Is Cernunnos the same as the Horned God of modern witchcraft?

They are connected, but a word count does not date the birth of the underlying idea as neatly as it first appears.

Margaret Alice Murray was an Egyptologist who wrote two influential books arguing that the European witch trials had persecuted the survivors of an old pagan religion. The first, The Witch-Cult in Western Europe, appeared in 1921. It already devotes a chapter called The God to the witches' supposed pre-Christian deity, and its preface says that part of the book had appeared earlier as an article titled The God of the Witches. In the full text, however, Cernunnos does not occur once and the exact phrase horned god does not occur once. The word horned appears once.

Her second, The God of the Witches, appeared in 1931. Its opening chapter is titled The Horned God. The phrase occurs 52 times, and Cernunnos appears 5 times, now presented as one of the greatest gods of Gaul and identified with Herne the Hunter and with a figure in Palaeolithic cave art.

The comparison therefore shows a change in Murray's label and archaeological synthesis, not the sudden invention of the whole idea between two books. By 1931 she had recast a deity already central to her 1921 argument as the Horned God and placed Cernunnos inside that synthesis. Gerald Gardner later drew heavily on Murray's framework when constructing Wicca, but that is a twentieth-century line of influence, not an unbroken tradition from ancient Gaul.

Murray's method is not accepted by historians or folklorists. It selected trial material to build a uniform hidden religion and explained away details that did not fit. Her Cernunnos has a smaller and more checkable problem too: she writes, more than once, that the Romans called him Cernunnos. The pillar was paid for by a Gaulish boatmen's guild, and the name on it is Gaulish. The direction of naming in the standard modern account runs backwards.

None of which makes the modern Horned God false to the people who honor him. A tradition roughly a century old is still a tradition. It is simply a different claim from an unbroken line back to Iron Age Gaul, and only one of the two can be checked.

Is Cernunnos the same as the Green Man?

No, and the difference is not horns. It is composition.

Green Man is a modern name for a varied group of foliate heads in medieval architecture. In the common form, a face is surrounded by foliage or sprouts leaves from its mouth. That is not the same composition as the complete seated, antlered figure on the Gundestrup plate, which holds a torc and a serpent and is surrounded by animals. The useful distinction is the composition, not a claim that every foliate figure lacks a body or accompanying objects.

The two figures do share one thing, and it is the thing this article keeps running into. The Green Man's name is modern as well. Lady Raglan applied it to foliate heads in a 1939 article in Folklore, beginning with a carving at Llangwm in Monmouthshire, and argued that the motif preserved a pagan fertility figure. Later folklorists accepted the convenient name without accepting that interpretation. Two familiar names applied to old European images, and both reached general use in the twentieth century with an interpretation already fitted.

The Cernunnos relief is ancient and the foliate-head carvings are medieval. The confident identification of both as named pagan figures is modern.

From the archive

Our own antlered figures are drawn as what the evidence actually supports: a forest presence surrounded by animals, without a caption claiming to know his job. Cernunnos: Runes and Laurel sets him in a ring of runes and laurel, and Horned God Cernunnos puts him among pines with a stag and a serpent, the same company he keeps on the Gundestrup plate. Cernunnos: Lord of the Wolves and The Horned God: Oak Crown and Wolves take the lord-of-the-animals reading at its word.

The rest of the old gods are in the Folklore collection and in Pagan Shirts. If you like a god whose sources argue with each other, the leshy is the Slavic forest spirit who walks travelers in circles, and Odin's ravens are a case where the texts survive and the retellings still drift. For an ancient deity with a much fuller textual record and a long modern afterlife, see Hecate.

Sources

• Alexandre Léonet, "Cernunnos: approche historiographique d'une figure emblématique gauloise," in Imagination et construction mentale. La fabrique du discours scientifique, Ausonius Éditions, collection Schol@ 1, 2022, 47 to 61. Supports the single labelled antlered image, the three competing explanations for its disputed initial, the related but unproven names Cerunincus, Carvonia, Jupiter Cerneni and Carnonos, and the sequence of scholarly identifications from Bacchus to Pan to hunting god to Dis Pater. Read it
• Phyllis Fray Bober, "Cernunnos: Origin and Transformation of a Celtic Divinity," American Journal of Archaeology 55, no. 1 (1951): 13 to 51. The foundational study of the Gallo-Roman material and the source for the fertility and abundance reading. JSTOR record
• Musée de Cluny, "Pillar of the Boatmen." Supports the monument's date, its discovery beneath Notre-Dame, and the mixture of Roman and Gaulish figures. Museum record
• The National Museum of Denmark, on the Gundestrup Cauldron. Supports the 1891 discovery in Rævemosen, the uncertainty about where the cauldron was made, and the museum's own description of its figures as unknown gods in a foreign style. Museum page
• Margaret Alice Murray, The Witch-Cult in Western Europe, Oxford, 1921. Searched in full: a chapter titled The God and a preface naming her earlier article The God of the Witches, but no occurrence of Cernunnos and no occurrence of the exact phrase horned god. Full text
• Margaret Alice Murray, The God of the Witches, 1931. Searched in full: opening chapter titled The Horned God, 52 occurrences of the phrase, 5 of Cernunnos, and the claim that the Romans called him Cernunnos and that the name passed into English as Herne. Book record
• Jacqueline Simpson, "Margaret Murray: Who Believed Her, and Why?" Folklore 105, nos. 1 to 2 (1994): 89 to 96. Supports the rejection of Murray's method and her influence on Gardner's Wicca. Journal article
• Lady Raglan, "The 'Green Man' in Church Architecture," Folklore 50, no. 1 (March 1939): 45 to 57. Applied the name Green Man to foliate heads and argued for a pagan fertility survival. Journal article
• Stephen Winick, "Green Man Connections: The Foliate Head," Library of Congress Folklife Today, 2023. Supports the variety of foliate-head forms and the distinction between the medieval carvings and later interpretations attached to them. Library of Congress
• Brandon S. Centerwall, "The Name of the Green Man," Folklore 108, nos. 1 to 2 (1997): 25 to 33. Supports the distinction between the useful modern name and Raglan's unsupported interpretation. Journal article
• Musée Saint-Remi, Reims, "Autel de Cernunnos," inventory 978.20189. Describes the antlered figure between Apollo and Mercury and the bag spilling seeds or coins between a stag and a bull. Museum record
• Ceisiwr Serith (David Fickett-Wilbar), "Cernunnos: Looking a Different Way," author's web draft. Supports the attributed interpretation of mediation, wealth, and exchange rather than an attested ancient function. Read the web draft
• Fred C. Woudhuizen, "Luwian Kuruntas and Celtic Cernunnos: Two Closely Related Manifestations of the Same Indo-European God," Živa Antika 64 (2014): 55 to 60. The argument that the Luwian stag-god and Cernunnos are one god under two names. Journal PDF
• Calvert Watkins, "A Celtic Miscellany," in Proceedings of the Tenth Annual UCLA Indo-European Conference, Los Angeles 1998, Journal of Indo-European Studies Monograph Series 32, Institute for the Study of Man, 1999, 3 to 25. The earlier linguistic work Woudhuizen builds on, cited by him at pages 14 to 20 for the Indo-European roots and for the Schimmel rhyton, Paris and Gundestrup images. Recorded here through Woudhuizen's citations; the original was not accessed.
• eDIL: Electronic Dictionary of the Irish Language, entry for benn, senses (a), (d), and (e). Supports the range of meanings including peak, animal horn, and point; it does not decide the meaning of the Dagda's name in its narrative context. Dictionary entry
• Maxim Fomin, "The Hunt Scenes of the Irish High Crosses and the Iconography of the Continental Celtic Stag Deity," Text and Image: Essential Problems in Art History 2 (2016): 6 to 18. Pages 11 to 12 discuss comparisons between Cernunnos and the Anatolian stag deity, drawing on linguistic and iconographic evidence. Full text
• Doris Srinivasan, "The So-Called Proto-Śiva Seal from Mohenjo-Daro: An Iconological Assessment," Archives of Asian Art 29 (1975/1976): 47 to 58. Supports Marshall's original reading, the Vedic sense of pashupati as a guardian of domestic animals, the four wild animals on seal 420, and the divine bull-man proposal. JSTOR record
• Royal Shakespeare Company, "Herne the Hunter," notes on The Merry Wives of Windsor. Supports the play's date of about 1597 and the statement that the earliest written account of Herne is the play itself. Play notes
• J. A. MacCulloch, The Religion of the Ancient Celts, Edinburgh, 1911. Supports d'Arbois de Jubainville's pairing of Cúchulainn and Conall Cernach with Smertrios and Cernunnos, and MacCulloch's criticism of that identification, including his discussion of the epithet Cernach. Full text
• Cath Maige Tuired: The Second Battle of Mag Tuired, edited and translated by Elizabeth A. Gray, Irish Texts Society, 1982. Supports the Dagda's answer Fer Benn and the woman's reply in the carrying scene. CELT text
• Kuno Meyer, ed., "Finn and the Man in the Tree," Revue Celtique 25 (1904), from Trinity College Dublin MS H 3 18. Supports Derg Corra in the treetop with the blackbird, the trout in the bronze vessel, and the stag. CELT text