Skip to content

Roman religion

Mithraism

Hundreds of rooms, one image repeated everywhere, and not a line of its own scripture. A religion known almost entirely from its architecture.

1st–4th century CE

What is attested

The Roman cult of Mithras is known from a large corpus of monuments and inscriptions. M. J. Vermaseren's mid-twentieth-century Corpus Inscriptionum et Monumentorum Religionis Mithriacae runs to some 2,400 entries and has been substantially added to since; published counts of 'Mithraic monuments' vary widely with what is being counted. What the corpus does not contain is a single sentence of Mithraic scripture, a single myth told by an initiate, or a single explanation of the cult's central image by anyone who belonged to it. This is not a gap that further excavation is likely to close in kind: the cult appears never to have produced, or at least never to have circulated, a written theology. Everything said about Mithraic belief is inference from buildings, images and dedications, supplemented by a handful of hostile or philosophically motivated outsiders.

The buildings are the most eloquent evidence. A mithraeum is a small rectangular hall, typically sunk below ground level or built to feel as though it were, with continuous masonry benches running along both long sides and a niche at the far end holding the cult relief. Capacity is characteristically a few dozen at most. Mithraea are consistently small and consistently numerous where the cult was strong — Ostia alone has produced well over a dozen. Whether that reflects deliberate subdivision, independent foundations by separate patrons, or successive periods of use is not recoverable from the buildings. Distribution correlates strongly with the Roman army and the machinery of imperial administration: the Rhine and Danube frontiers, Rome and its port, Britain, the eastern garrisons. The military weighting varies by region, and work on the dedicators has shown the civilian component at Rome and Ostia to be large. The chronology runs from the late first century CE to the late fourth. The mithraeum at Dura-Europos on the Euphrates, built by Palmyrene archers in Roman service, carries dedicatory inscriptions dated to 168 and 171 CE and is among the most securely dated examples.

The cult image is astonishingly stable across that whole range. It shows a young man in eastern dress and a Phrygian cap kneeling on the back of a bull and driving a dagger into its shoulder, his head characteristically turned away from the act. Around him recur a dog and a snake reaching toward the wound, a scorpion at the bull's genitals, a raven, and two torchbearers dressed like the central figure, one holding his torch up and one down; busts of the Sun and Moon occupy the upper corners, and the whole is often set within a cave-like frame or an arch of zodiacal signs. Variants add episodes: a figure emerging from rock, a banquet shared with the Sun, a bow drawn at a rock face. That this repertoire recurs from the Euphrates to the Tyne with such consistency implies transmitted content of some kind. What that content was, no ancient source states.

The internal organisation is better documented than the doctrine. The full seven-grade list survives in one ancient text, a letter of Jerome written in 403 CE: raven, then a second grade whose name is disputed — the manuscripts have been read both as nymphus (bridegroom) and as cryphius (the hidden one), and the painted inscriptions at Santa Prisca support nymphus — then soldier, lion, Persian, sun-runner, father. Archaeology corroborates the seven-fold structure but not the list exactly — the second grade appears at Santa Prisca as nymphus, bridegroom, and the divergence from Jerome is itself a standing problem. Porphyry names individual grades, lions and ravens, in a separate passage. The floor mosaic of the Mithraeum of Felicissimus at Ostia lays out seven panels of symbols along the central aisle between the benches, and painted figures and inscriptions from the mithraeum beneath the church of Santa Prisca in Rome, excavated from the 1950s, show ranked members in procession. Dedications regularly record a man's grade and his relation to a pater, the senior figure of a given community. The named dedicators are soldiers, veterans, imperial slaves and freedmen, customs officials and traders — a distinctly male, mobile, service-oriented cross-section. No securely attested female initiate is known; the one literary text that complicates this, Porphyry's reference to women connected with the rites, is of disputed bearing.

Practice, as opposed to belief, leaves recoverable traces. The benches are dining benches: excavated mithraea yield animal bone, cooking and serving vessels, hearths and lamps in quantities consistent with regular communal meals. Water sources, wells and basins appear frequently enough to suggest ritual use. The Mainz mithraeum produced a large ceramic vessel, published in the 1990s, whose applied relief figures appear to show ritual action involving a seated figure, an archer and initiates in postures of submission — one of the very few objects that seems to depict what actually happened in the room rather than what was depicted on the wall. It remains an object requiring interpretation, and the interpretations differ.

How the rite worked

A mithraeum was a private construction, paid for by individuals whose names are recorded on the dedications: a pater, a patron, occasionally a group of soldiers acting collectively. The Walbrook mithraeum in London produced a marble relief of the bull-slaying dedicated by Ulpius Silvanus, described as a veteran of the Second Augustan legion, who records having been made — either discharged as a veteran or, on the reading many prefer, initiated — at Arausio in southern Gaul. That single stone shows the mechanism of the cult's spread: a man moves with the army, is enrolled somewhere along the way, and pays for the furnishing of a room at the far end of his career. Fittings were bought, altars dedicated, reliefs commissioned locally or imported, and worn-out mithraea were repaired and re-dedicated with inscriptions recording the fact.

Members advanced through grades, but how the grades functioned is disputed. The Ostia mosaic and the Santa Prisca paintings show the seven-fold scheme as an ordered sequence with distinct symbols, and several sources associate the grades with the planets. Whether every member was expected to climb the whole ladder, or whether the higher grades constituted a functional priesthood within a congregation whose ordinary members remained at a lower level, is argued from the relative frequency of grade titles in inscriptions and is not settled. What is clear is that the pater was the ranking figure of a given community and that grade was worth recording publicly on stone.

Initiation involved ordeal and role-play of some kind. Tertullian, writing around 200 CE as a hostile Christian, describes a Mithraic soldier being offered a crown at sword-point in a dark cave and required to refuse it, declaring Mithras his crown. Painted lines beside the ranked figures at Santa Prisca appear to be ritual acclamations, and the Mainz vessel seems to show a kneeling, blindfolded or bound figure confronted by an archer. Several mithraea contain pits, niches and installations that excavators have interpreted as ordeal or concealment features. These are consistent with an initiation involving darkness, physical trial and a scripted refusal, but they do not amount to a reconstructed rite, and the sequence, the words spoken and the meaning attached are unrecovered.

The core regular activity was a meal taken reclining on the two benches, facing the cult relief and each other. The meal is visible in the debris — bone, cooking and serving vessels, hearths — but its scale and cost are not recoverable, and the assemblages vary by site. Some mithraea show deliberate decommissioning in the fourth century — images defaced, altars overturned, deposits sealed — but whether a given case reflects Christian violence, official closure, or an act by the members themselves has to be argued site by site and often cannot be decided.

The evidence and its limits

Each class of evidence below is followed by what it cannot show. That second half is not a disclaimer; it is the part most accounts of ancient religion leave out.

  • Archaeology

    Excavated mithraea across the empire supply the plan, the seating, the orientation toward the cult image, the evidence for dining, and in some cases the abandonment sequence.

    What it cannot show — A building tells you the shape of an assembly and roughly how many attended; it cannot tell you what was said, believed, promised or taught in it, nor how the room's arrangement was interpreted by those who used it.

  • Images

    The tauroctony and its subsidiary scenes constitute the cult's principal surviving statement about itself, and their consistency across two centuries and thousands of miles is itself a datum.

    What it cannot show — The images carry no captions and no accompanying narrative. Any story reconstructed from them is a modern hypothesis about a lost text, and the same image is compatible with several mutually exclusive readings.

  • Inscriptions

    Dedications name members, grades, offices, occupations, military units and occasionally dates, giving the cult's social profile and chronology.

    What it cannot show — Dedications record status and generosity, not doctrine. They also over-represent those who could afford stone, and their silence about women and about the poor may reflect the medium as much as the membership.

  • Literary sources

    Porphyry, Tertullian, Jerome and a scatter of poets provide the only verbal statements about the cult that survive from antiquity, including the sole list of the seven grades.

    What it cannot show — Every one of these writers is an outsider — a Neoplatonist reading the cult allegorically, a hostile Christian polemicist, a churchman recording its destruction. None describes a rite he attended as a member, and their interests actively shape what they report.

What the sources do not record

  • There is no Mithraic myth. Not a fragment of narrative survives from within the cult explaining who Mithras was, what the bull was, why it was killed, or what the killing accomplished. The absence is the single most important fact about the evidence.

  • There is no liturgy, prayer, hymn or ritual instruction from the cult. What was said in a mithraeum is entirely unrecovered.

  • No Mithraic community's membership record survives, and no internal rule or constitution. We do not know how a man was recruited, what he was told before joining, or what he paid.

  • No woman is securely attested as an initiate. Porphyry refers to women connected with the rites under an animal designation, but whether that passage describes participation in the mysteries is disputed. Whether the absence reflects a formal exclusion, a social fact about the milieux in which the cult spread, or a bias in the surviving media cannot be determined from the evidence itself.

  • Nothing indicates any central authority, canon, or mechanism of coordination across the empire — yet the iconography is highly uniform. The evidence records the uniformity and is silent about how it was maintained.

What the evidence supports

Evidence level: Probable

Seven panels of symbols laid along the aisle between the benches, corroborating a seven-fold scheme.

Floor mosaic of the seven grades, Mithraeum of Felicissimus (CIMRM 299) (Ostia). The assignment of each panel to a named grade is an interpretation informed by Jerome and by the Santa Prisca inscriptions, not a reading of labels on the mosaic itself.

Evidence level: Probable

Ranked members depicted in procession with grade attributes, accompanied by painted lines that appear to be ritual acclamations — among the closest approaches we have to the cult's own words.

Painted figures, processions and inscriptions in the mithraeum beneath the church of Santa Prisca, excavated by Vermaseren and van Essen from the 1950s and published in 1965 (Aventine, Rome).

Evidence level: Documented

An individual soldier's career carried the cult across the empire, and that the dedicator recorded where he had been made — a rare glimpse of a member's personal trajectory through the cult.

Marble tauroctony relief dedicated by Ulpius Silvanus, veteran of legio II Augusta, from the Walbrook mithraeum excavated in 1954 (London (Londinium); London Museum collections).

Evidence level: Probable

A securely dated mithraeum at the eastern edge of the empire, founded by a specific military unit, with successive building phases and painted decoration.

Mithraeum with dated dedicatory inscriptions of 168 and 171 CE, built by Palmyrene archers in Roman service (Dura-Europos, on the Euphrates; wall paintings now Yale University Art Gallery).

Evidence level: Probable

Two scenes involving seven figures, including a seated figure and an archer confronting a kneeling initiate, widely read as depicting ritual action rather than myth — unusual and much-discussed evidence for what happened in the room.

Large ceramic 'snake vessel' with applied relief figures, recovered in 1976 from unsupervised construction work on the Ballplatz site and restored in 1980-81, published in 1994 (Mainz (Mogontiacum)).

Stories told about the rite

Ancient writers offered these as explaining the practice. They are evidence for what the tradition said about itself, and they are not the origin of anything: each is given with the author who tells it and the distance between the two.

  • Evidence level: Mythological

    Porphyry, On the Cave of the Nymphs, third century CE — at least two centuries after the earliest Mithraic monuments.

    Porphyry, drawing on a writer he names as Eubulus, reports that Zoroaster was the first to consecrate a natural cave to Mithras in the mountains of Persia, and that the cave was made an image of the cosmos Mithras had fashioned.

    Offered as explaining — Why Mithraic worship takes place in caves or cave-like rooms, and why the mithraeum's decoration has cosmic elements.

    This is a Neoplatonist philosopher explaining a contemporary cult in terms congenial to his own allegorical method, and attributing its origin to Zoroaster gives it the antiquity his argument wants. It is evidence for how a learned third-century outsider understood mithraea. It is not evidence that the cult originated in Persia or that Zoroaster had anything to do with it.

  • Evidence level: Mythological

    Plutarch, Life of Pompey, written around the turn of the second century CE, about the events of the 60s BCE.

    Plutarch reports that the Cilician pirates suppressed by Pompey performed strange rites, including those of Mithras, which he says they transmitted onward.

    Offered as explaining — How a Persian-named god's worship arrived in the Roman west, and dates that arrival to the first century BCE.

    A single sentence in a biography, written well over a century after the events, is the entire basis for the pirate-origin story. No Mithraic monument is dated anywhere near that early, and nothing links the archaeological cult to Cilicia. It is a report about the cult's origin, not a record of it.

  • Evidence level: Mythological

    Iconographic, second to fourth centuries CE, from many sites.

    Reliefs and statues show a figure emerging from a rock, sometimes flanked by torchbearers and encircled by a serpent; the type is conventionally identified as the birth of Mithras from the rock.

    Offered as explaining — Where Mithras came from — an origin without parents, tied to the stone and the cave.

    The identification as a birth is a modern reading of a recurring image type, plausible but derived from the picture itself and from a very few late labels. No ancient narrative of Mithras's birth survives, so this is a story we reconstruct from the tradition's images rather than one the tradition tells us.

Vocabulary

mithraeum
The cult building: a small, low, rectangular hall with benches along both long sides and the cult image at one end, built or fitted to evoke a cave. Capacity is typically a few dozen.
tauroctony
The bull-slaying image, with its fixed cast of subsidiary figures, set in or above the niche at the end of the hall. It is the standard cult image wherever one survives, though many identified mithraea preserve none. It is the cult's central statement about itself and its meaning is not explained by any ancient source.
pater
'Father', the highest of the seven grades and the ranking figure of a given congregation. Members frequently record their relation to a named pater in dedications.
Cautes and Cautopates
The two torchbearers flanking the bull-slaying, dressed as Mithras is, one holding his torch upward and one downward. Their function is inferred from the symmetry of the image; no ancient text explains them.
CIMRM
Corpus Inscriptionum et Monumentorum Religionis Mithriacae, Vermaseren's catalogue of Mithraic material published 1956–60. Monuments are conventionally cited by CIMRM number.
tauroctony variants / side scenes
Additional episodes carved around the main relief — a figure rising from rock, a banquet with the Sun, an archer at a rock face. They are clearly episodic and clearly refer to something narrative, but the narrative itself is lost.

Contested points

Evidence level: Disputed

Is the Roman cult a continuation of Iranian Mithra-worship, or a Roman creation that borrowed a Persian name and costume?

Franz Cumont's reconstruction, dominant from the turn of the twentieth century, treated the Roman mysteries as Iranian religion transmitted westward, and read the tauroctony as an Iranian creation myth. That framework was severely criticised at the First International Congress of Mithraic Studies at Manchester in 1971, where Iranianists pointed out that the reconstruction finds no support in the Iranian material and in places contradicts it, and that no Iranian text supplies a bull-slaying of the required kind. The alternative position holds that the Roman cult was substantially invented in the Roman world, deliberately dressed in Persian imagery for the prestige of antique eastern wisdom. A middle position accepts genuine Iranian elements in the name, dress and some motifs while denying that the cult's content can be reconstructed from Iranian sources. The debate is not closed, but the wholesale Cumontian transfer is no longer the default.

Evidence level: Disputed

What does the tauroctony mean?

One family of readings is astronomical: the recurring figures correspond to constellations lying along a band of the sky, and the image encodes a celestial configuration. David Ulansey argued specifically that Mithras represents Perseus and that the image commemorates the precession of the equinoxes; Roger Beck developed a more elaborate account in which the mithraeum and its imagery constitute a 'star-talk' that members learned to read. Others hold that the astronomical correspondences are real but decorative or secondary, and that the image's core is sacrificial or soteriological; still others, notably in the sceptical tradition, hold that in the absence of any explanatory text the honest position is that we do not know, and that the elegance of a hypothesis is not evidence for it.

Evidence level: Disputed

Were the seven grades an initiatory ladder every member climbed, or a priestly hierarchy within a wider membership?

The initiatory reading takes the Ostia mosaic's sequential layout and Jerome's list at face value: a member entered as raven and advanced, with the pater at the summit. The alternative, argued in recent scholarship, notes that grade titles appear in inscriptions with a distribution that would be odd if every member held one, and proposes that the named grades constituted an officiating hierarchy while most participants held no grade at all. The evidence relevant to deciding this is inscriptional frequency and iconographic layout, both of which are open to more than one statistical reading.

Where this comes from

  • Jerome, Letter to Laeta (Epistulae) 107

    Cites the case of a relative of the addressee who, as urban prefect, destroyed a Mithraic cave and its images, and in doing so lists the names of the grades through which initiates passed — raven, cryphius, soldier, lion, Persian, sun-runner, father. The second name diverges from the nymphus attested epigraphically at Santa Prisca, and the discrepancy is unresolved.

  • Porphyry, On the Cave of the Nymphs (De antro nympharum) 6

    Explains that a cave was consecrated to Mithras as an image of the cosmos he had made, and attributes the practice of using caves to Zoroaster, citing an earlier writer.

  • Porphyry, On Abstinence from Killing Animals (De abstinentia) 4.16

    Refers to participants in the rites of Mithras called lions and ravens, and to women associated with the rites under an animal designation of their own; the passage is the only literary evidence bearing on female participation and its meaning is contested.

  • Tertullian, De corona 15

    Describes, disapprovingly, an initiation in a dark cave in which a crown is offered to the candidate on a sword point and he is instructed to push it away and declare Mithras to be his crown; Tertullian's point is that even this cult refuses the garland Christians should refuse.

  • Plutarch, Life of Pompey 24

    Reports that the Cilician pirates practised secret rites, among them those of Mithras, which continued to be performed after their suppression.

  • Statius, Thebaid

    Contains a brief simile referring to Mithras wrestling with horns in a Persian cave, generally taken as the earliest surviving Latin literary mention of the god in a form recognisably close to the cult image.

Covered elsewhere

How to read the evidence labels

Evidence level: Documented
Supported by a named primary text or by excavated material evidence.
Evidence level: Probable
Supported by strong inference that most specialists accept.
Evidence level: Disputed
Substantial specialist disagreement exists; competing readings are stated.
Evidence level: Literary
Belongs to poetic or mythological construction rather than to history.
Evidence level: Mythological
Belongs to shared religious or mythic tradition rather than to any single text.
Evidence level: Unknown
The available evidence is insufficient to decide.

These labels describe the status of the evidence, not our confidence in a conclusion. A claim marked Literary is not a claim we doubt; it is a claim about what kind of thing a poem is.