Skip to content

Rite and practice

Animal sacrifice

The central act of Greek religion, and the best-documented thing about it is what it cost. The calendars are budgets.

Bronze Age – 4th century CE

What is attested

The best-documented thing about Greek animal sacrifice is not what it meant but what it cost and when it was due. Attic demes and civic bodies inscribed sacrificial calendars on stone: month-by-month schedules naming the recipient deity, the species and often the sex, age or colour of the victim, a ceiling price in drachmas, and sometimes who was entitled to eat and who was to be paid. The calendar of the deme Thorikos (SEG 33.147) preserves all twelve months in the standard Athenian order; the calendars of Erchia (SEG 21.541) and of the Marathonian Tetrapolis (IG II² 1358, re-edited as SEG 50.168) are comparable documents. Read as what they are — budgets, ratified and displayed — they establish that sacrifice was a recurring, priced, audited municipal obligation, planned a year in advance, and that a small Attic community expected to kill animals for named gods and heroes on dozens of fixed days. Lysias' prosecution speech against Nikomachos, who was charged with mishandling the re-inscription of Athens' sacred calendar, turns on the same premise: adding sacrifices to the list meant money, and money spent on new rites was money not spent on inherited ones.

The physical residue confirms the practice independently of any text. At the ash altar of Zeus on Mount Lykaion in Arcadia, micromorphological analysis has shown that the mound is not soil but pulverised burnt animal bone mixed with ash, carbon and potsherds, accumulated directly on bedrock; roughly 98 per cent of the identified faunal remains are burnt sheep or goat thigh bones, with much smaller quantities of pig and cattle, and burning at the spot is documented from about the sixteenth century BC, at the beginning of the Mycenaean period, and continuing into the Hellenistic period, though the terminal date is given variously in the excavation reports. Two things follow that no literary source is needed to establish: a narrow and repeated selection of animal parts was burnt in the same place over an extraordinary span of time, and the whole animal was not consumed by fire. Which anatomical parts should be identified with the god's portion of the literary descriptions is itself argued over — the Homeric formula speaks of thigh bones wrapped in fat, while the burnt material recovered from altars has been read both as femur and as tail and sacrum — but the fact of a restricted, repeated selection is not in doubt. Whatever else sacrifice was, it produced a residue of selected burnt bone and a great deal of edible meat that went somewhere else.

Where the meat and the by-products went is also documented. Sacred laws specify priestly perquisites down to particular cuts — a leg, the skin, a portion of the entrails — as an enforceable entitlement rather than a courtesy. The Athenian dermatikon accounts (IG II² 1496) record the state's revenue from the sale of hides taken from public sacrifices, itemised by festival across the 330s BC, including the Dionysia in Piraeus, the City Dionysia, sacrifices to Zeus Soter and the Asklepieia. A sacrificial animal was therefore a divided asset: an incinerated portion for the god, defined perquisites for the officiant, hides converted into cash for the treasury, and meat distributed to participants. This is an attested economy, not an inference from myth.

Iconography supplies the choreography. Attic black- and red-figure vases and votive reliefs — the corpus assembled by F. T. van Straten in Hierà kalá (1995) is the standard reference — show a repeating sequence: the animal led on a rope in procession, a basket-bearer carrying the kanoun, a water vessel, the burning altar, and, after the killing, attendants roasting the splanchna on long spits beside the altar while the god's portion burns. What the painters show consistently is the moment before the blow and the moment of the roasting; the killing itself is largely absent from the Attic repertoire. That absence is a genuine feature of the evidence and should be reported as such rather than explained away.

The literary evidence is real but must be dated. The Homeric poems describe sacrificial procedure in some detail — the hecatomb for Apollo at Chryse in Iliad 1, Nestor's sacrifice at Pylos in Odyssey 3, with barley grains, the burning of thigh-portions wrapped in fat, libations of wine and the roasting of entrails on spits — and the outline matches what calendars, altars and vases attest centuries later. But these are hexameter poems composed for performance, not manuals, and they idealise: gilded horns and an on-call goldsmith belong to an epic register. At the other end, Porphyry's De Abstinentia (2.29–30), written in the third century AD by a philosopher arguing against eating meat, preserves earlier material traced to Theophrastus; it is a valuable but tendentious witness, reporting a classical Athenian rite at a distance of some seven centuries and through an argument he wants to win.

How the rite worked

What follows is a modern reconstruction, and it should be read as one: no single ancient source describes the sequence entire, and the material from which it is assembled is weighted heavily towards Attica and towards Homer. A public sacrifice is presented as beginning with a procession to the sanctuary. The Attic vases and reliefs collected by van Straten show the animal led on a rope, decorated, and a participant carrying the kanoun, a flat basket holding barley grains, fillets and the knife concealed beneath them. Homer and Attic drama supply the acts at the altar: the washing of hands, the scattering of barley grains, the cutting of hair from the victim's forehead to be thrown on the fire, and a prayer naming the god and the request. The killing is described in texts rather than shown on pots — small victims held over the altar, larger ones brought down first, the throat cut, the blood directed onto or around the altar. The ololygē, the ritual cry raised by women at the moment of the kill, is attested in Odyssey 3 and in Aeschylus. Each element has its own witness; the continuous sequence is the work of modern scholarship on an Attic-weighted corpus.

Butchery followed immediately and was technical work. The god's portion — thigh bones wrapped in fat, with other selected pieces and often the tail — went onto the altar fire with wine poured over it; the splanchna, the heart, liver, lungs and kidneys, were spitted and roasted quickly and eaten at the altar by the inner circle of participants. The remaining meat was boiled or roasted and distributed. Distribution was the politically sensitive part: sacred laws and decrees specify who receives portions, in what order, and whether shares are equal or graded, and public sacrifices funded by a city were frequently framed as meat for the citizen body.

Personnel and money were formalised. A priest or priestess of the particular cult presided and drew defined perquisites; a mageiros — at once butcher, cook and sacrificial technician — could be hired for the killing and carving; magistrates such as demarchs or hieropoioi supervised expenditure and answered for it. The calendars set a maximum price per victim, and the relative order of magnitude is clear even where individual figures are uncertain: cattle cost several times what sheep and goats cost, and piglets were cheap enough for small groups and modest occasions. Hides were sold, and in Athens the proceeds were accounted for centrally.

Timing was calendrical: for most rites the day was prescribed rather than chosen, and only a few texts specify a time of day. Some documents distinguish forms of offering by verb, by altar type, and by whether the meat could be carried away from the sanctuary or had to be consumed on the spot — a real distinction in the texts, though the neat modern division of all Greek sacrifice into 'Olympian' and 'chthonian' types imposes more system on the evidence than the evidence carries.

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.

  • Inscriptions

    Sacrificial calendars, sacred laws and financial accounts give species, sex, price, date, recipient deity, perquisites and revenue — the hard skeleton of the practice.

    What it cannot show — They are prescriptive and financial. They record what was budgeted and owed, not whether any particular sacrifice took place as written, and they say nothing about what participants believed. The corpus is also overwhelmingly Attic, so Athenian practice is far better documented than anywhere else.

  • Archaeology

    Ash altars and burnt-bone deposits show which parts of which animals were repeatedly burnt at a fixed spot, and for how many centuries.

    What it cannot show — Bone cannot name the deity, recover the words spoken, or distinguish a civic festival from a private offering. Preservation favours burnt and durable material over meat, blood, cakes and textiles, and excavation and publication of faunal material is uneven between sanctuaries, so intersite comparison is fragile.

  • Images

    Attic vases and votive reliefs preserve the sequence of procession, altar, kanoun and the roasting of entrails, and show equipment that no text describes clearly.

    What it cannot show — Vases were made as tableware for drinking parties and follow pictorial convention; they select, compress and repeat. They cannot be read as reportage, and their systematic avoidance of the moment of killing is a choice by painters, not a record of what happened.

  • Literary sources

    Homer, Attic drama and later antiquarian writers preserve procedural detail, terminology and the framing prayers that documents omit.

    What it cannot show — Every one of these texts was written for effect — epic performance, comic laughter, philosophical polemic — and most are separated from the practice they describe by centuries in one direction or the other. Porphyry writes in the third century AD, some seven and a half centuries after the classical rite he reports, drawing on Theophrastus, who is himself roughly six centuries earlier than Porphyry — and he is arguing against animal sacrifice as he does so.

What the sources do not record

  • The words spoken at a sacrifice are not preserved. There is no surviving liturgical text for an ordinary Greek sacrifice — no fixed prayer, no formula that we can be sure was said at the altar. What we have are prayers composed inside poems.

  • Household and small-scale offering is nearly invisible. Cakes, grain, incense, first-fruits and a piglet at a domestic hearth left almost no inscribed or archaeological trace, so the surviving record systematically over-represents state-funded animal killing and under-represents everything cheaper.

  • The calendars have no counterpart in performance records. Nothing tells us that a scheduled sacrifice was actually carried out, that a victim was ever rejected as unsuitable, or that a rite lapsed for lack of funds — only that it was voted and priced.

  • Non-Athenian practice is thinly documented by comparison, and the great sanctuaries of the Greek east, Sicily and the Black Sea are represented by scattered documents rather than continuous series.

  • What participants thought the burning fat accomplished is not recorded by any participant. Every ancient explanation we possess comes from poets, philosophers or antiquarians who were, in a sense, outsiders explaining a routine to readers.

What the evidence supports

Evidence level: Probable

Direct physical proof of repeated burnt animal offering at a fixed point over a span measured in millennia, with a strong preference for small stock at this sanctuary, and confirmation that only part of the animal was burnt.

Ash altar of Zeus, with faunal and micromorphological analysis: the mound consists of pulverised burnt bone, ash and carbon on bedrock; c. 98% of identified faunal remains are burnt sheep or goat thigh bones, with lesser pig and cattle; burning attested from c. the 16th century BC, at the beginning of the Mycenaean period, and continuing into the Hellenistic period, the terminal date being reported variously in the excavation publications (Mount Lykaion, Arcadia (Mt. Lykaion Excavation and Survey Project)).

Evidence level: Documented

A complete deme year of prescribed offerings in standard month order, naming recipients including local heroes and specifying victims, and recording an obligation to send an offering for Poseidon to Sounion.

Sacrificial calendar of the deme Thorikos, inscribed stone (SEG 33.147) (Thorikos, Attica).

Evidence level: Documented

State revenue from the sale of hides of publicly sacrificed animals, itemised by festival across the 330s BC, demonstrating that sacrificial by-products were centrally accounted for as civic income.

Dermatikon accounts of the treasurers of Athena (IG II² 1496) (Athens).

Evidence level: Probable

The standard visual sequence of a sacrifice, the equipment used, and the conspicuous rarity of the killing itself in the Attic repertoire.

Attic black- and red-figure vases and votive reliefs showing procession, kanoun-bearer, blazing altar and attendants roasting splanchna on spits (Attica and export markets; corpus assembled in van Straten, Hierà kalá (1995)).

Evidence level: Disputed

A famous literary description of an ash altar that has not been securely matched to an excavated feature — a case where the text is richer than the ground.

The great ash altar of Zeus described by Pausanias (Olympia).

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

    Hesiod, Theogony, composed in hexameter probably around 700 BC.

    At Mekone, Prometheus divided a great ox and set out two portions: the edible meat and innards concealed inside the unappetising stomach, and the bare bones dressed in glistening fat. Zeus chose the fat-covered bones, and from that day humans burn white bones on smoking altars for the gods and keep the meat.

    Offered as explaining — Why the gods receive bone and fat while the worshippers eat the flesh — the division of the sacrificial animal.

    This is a theogonic poem accounting for an arrangement that already existed when it was composed. Burnt bone had been accumulating on the Lykaion altar for some eight centuries before Hesiod wrote. The story explains the division as the outcome of a trick between Zeus and a Titan; it supplies neither a date nor a cause for the practice, and treating it as the historical origin of the meat division inverts the evidence.

  • Evidence level: Mythological

    Pausanias, Description of Greece 1.24.4 (second century AD); Porphyry, De Abstinentia 2.29–30 (third century AD, drawing on material traced to Theophrastus, later fourth century BC).

    An ox ate the offering cake from the altar of Zeus Polieus on the Athenian Acropolis; the man who struck it down fled or was tried, and in the ensuing inquiry the water-carriers accused the sharpeners of the axe, who blamed the smiths, until the axe itself was condemned as the guilty party and disposed of.

    Offered as explaining — The peculiar Athenian rite of the Bouphonia at the Dipolieia, in which responsibility for the killing was formally displaced onto the implement.

    A story told to explain an unusual local rite whose oddity clearly demanded explanation already in antiquity. Our fullest version comes from a vegetarian polemicist writing seven centuries or more after the classical rite. It documents how Athenians and their interpreters rationalised the rite, not why the rite exists.

  • Evidence level: Mythological

    Pausanias, Description of Greece, book 5 (second century AD), reporting local Elean tradition.

    The great ash altar of Zeus at Olympia was founded by a heroic figure at the origin of the games, and grew from the thigh-ash of every subsequent sacrifice.

    Offered as explaining — The antiquity and prestige of a specific altar, and why its ash was ritually maintained.

    A charter narrative in the service of a sanctuary's status, recorded by a traveller a thousand years after the period it claims to describe. It cannot date the altar; only excavation could, and the great ash altar Pausanias describes has not been securely identified on the ground.

Vocabulary

thysia
The standard form of Greek sacrifice: an animal killed at an altar, a selected portion burnt for the god, the remainder eaten by the participants.
splanchna
The internal organs — heart, liver, lungs, kidneys — roasted quickly on spits at the altar and eaten first, on the spot, by the immediate circle of participants.
mēria / osphys
The thigh bones and the lower back with the tail: the parts characteristically wrapped in fat and burnt as the god's portion.
kanoun
The flat ritual basket carried in procession, holding barley grains and fillets with the sacrificial knife concealed underneath.
ololygē
The ritual cry raised by women present at the moment the victim was struck.
mageiros
The sacrificial butcher-cook, a hired professional who killed, flayed, carved and cooked; a trade, not a priesthood.
dermatikon
The revenue accruing to the Athenian state from the sale of hides of animals sacrificed at public expense.
hiereion
The sacrificial victim itself, specified in calendars by species, sex, age and maximum price.

Contested points

Evidence level: Disputed

What was Greek animal sacrifice fundamentally about?

One line, running from Karl Meuli to Walter Burkert, derives the rite from Palaeolithic hunting and reads its central drama as guilt over killing, managed by a 'comedy of innocence' in which everyone disclaims responsibility — the Bouphonia axe-trial being the showpiece. A second, associated with Jean-Pierre Vernant and Marcel Detienne, treats sacrifice as an alimentary code: killing in order to eat, and in eating, defining the human place between beasts who eat raw and gods who consume smoke. F. S. Naiden (Smoke Signals for the Gods, 2013) rejects the premises of both, arguing that they overestimate the animal and underestimate the god, ignore the well-attested possibility that an offering was refused, and cannot explain why bronze figurines, cakes and incense functioned in the same way as slaughtered animals.

Evidence level: Disputed

Is the 'Olympian versus chthonian' division of Greek sacrifice a real ancient category or a modern schema?

The traditional handbook model sets a bright line between offerings to sky gods — daytime, raised altar, shared meal — and offerings to the dead, heroes and underworld powers — nocturnal, low hearth, victim burnt whole and not eaten. Against this, work on hero-cult ritual by Gunnel Ekroth and others shows that the ancient vocabulary (thyein, enagizein, bomos, eschara) is used far less consistently than the model requires, and that hero-cult sacrifices very often produced meat that was eaten; sceptics such as Scullion have argued in the other direction that the distinction, while imperfect, tracks something the Greeks did observe. The dispute matters because the model is routinely used to fill gaps in fragmentary sacred laws.

Evidence level: Disputed

How closely does the Homeric description correspond to real practice?

One view treats the Homeric sacrifice scenes as a reliable snapshot of eighth- or seventh-century procedure, since their sequence agrees with what later documents and vases attest. The other holds that epic sacrifice is a formulaic and idealising set-piece — gilded horns, a goldsmith on call, unfailingly willing victims — which preserves the outline of a real rite but not its ordinary texture, and which archaeology cannot corroborate detail by detail because the osteological record does not map neatly onto the poems' 'thigh bones wrapped in fat'.

Where this comes from

  • Hesiod, Theogony 535–560

    Narrates the division of an ox at Mekone, Prometheus' concealment of the good meat inside the paunch and of the bones inside fat, Zeus' choice of the bones, and the consequence that people burn bones for the immortals.

  • Homer, Iliad 1.447–474

    Describes a hecatomb for Apollo at Chryse: the arrangement of the animals, scattering of barley, the killing and flaying, the burning of thigh-pieces wrapped in fat with wine poured over them, the roasting of the entrails on spits, and the meal and singing that follow.

  • Homer, Odyssey 3

    Book 3 describes Nestor's sacrifice of a heifer at Pylos, including the summoning of a goldsmith to gild the animal's horns, the preparatory acts at the altar, the cry raised by the women, and the roasting of entrails before the wider meal.

  • Pausanias, Description of Greece 1.24.4

    Reports the killing of the ox at the Athenian Dipolieia and the subsequent displacement of blame from person to person and finally onto the axe, which was condemned.

  • Pausanias, Description of Greece 5.13.8–11

    Describes the altar of Zeus at Olympia as a mound built up from the ash of burnt thigh-portions, gives its measurements, and reports the annual procedure by which ash was carried up and mixed with river water to renew it.

  • Porphyry, De Abstinentia (On Abstinence from Animal Food) 2.29–30

    Gives an account of the Athenian ox-killing rite, including the sequence of accusations ending with the condemnation of the axe, within an argument against the sacrifice and eating of animals.

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.