Skip to content

Rite and practice

Asylum and supplication

Sacred ground could stop a killing. The interesting evidence is the cases where it did not, and what the city did afterwards.

Archaic Greece – Hellenistic period

What is attested

Two different things are usually run together under 'asylum', and the evidence for them is of different kinds. The first is hiketeia, personal supplication: a ritual posture — grasping a person's knees or chin, sitting at an altar or hearth, carrying a wool-wound branch — by which an individual placed themselves under the protection of Zeus Hikesios and made a public claim that the community then had to answer. The second is asylia, a formal inter-state status conferred by decree on a sanctuary, a city or an individual. Asylia is documented in bulk, by inscriptions; hiketeia is documented mainly by narrative, and mainly at the moments when it broke down.

The documentary institution is well recorded. From roughly the 260s BC to the first century BC, Greek cities sent sacred envoys, theoroi, to other cities, leagues and Hellenistic kings, requesting recognition that their territory or sanctuary was hiera kai asylos, sacred and inviolable, and the replies were inscribed together in a monumental dossier at home. Kent Rigsby's Asylia (1996) assembled the standard comprehensive corpus. The fullest case is Magnesia on the Maeander, and it took the Magnesians two attempts. A reported epiphany of Artemis Leukophryene in the 220s BC and a consultation at Delphi lay behind an initial approach that did not secure general recognition; the successful campaign came with the reorganised festival, first celebrated in 208 BC, when envoys went out to well over a hundred cities and courts. That the first attempt failed is itself good evidence that recognition was neither automatic nor cheap; the resulting decrees, together with a narrative inscription explaining the request (I.Magnesia 16), were cut on the sanctuary's walls. Teos obtained a comparable set of recognitions around 204/3 BC, including replies from the Aetolians, the Delphians, the Amphictyony, the Athamanians and a series of Cretan cities, in a process bound up with Antiochos III's dealings with the city. What these stones attest with certainty is procedure, ambition and expense: a city decided, sent, received, and displayed.

What asylia protected against was specific. Greek states recognised sylē — the right of reprisal, seizing the persons or property of citizens of a community that had failed to give redress — as a legitimate remedy, and it made travel and trade hazardous. Asylia was exemption from that. Cretan cities, whose treaties are our densest evidence for reprisal practice, are correspondingly prominent among the granting parties. On this reading the institution is a narrow one: an asylia decree is not a general declaration that no violence may occur at a sanctuary, but the removal of a particular legal liability, wrapped in the prestige of divine inviolability. How much practical force the grants carried is the first dispute below.

Personal supplication has a different evidentiary profile. Altars, hearths, sacred groves and the bounded temenos functioned as physical refuges, and Athenian practice included a formal channel by which a suppliant could place a bough on an altar and thereby put a petition before the council or assembly. But almost every supplication we can name is one that ended badly, because violations were what historians recorded and what politicians could use. Thucydides describes the followers of Cylon, besieged on the Athenian Acropolis and dying of hunger, seating themselves as suppliants at the altar, being brought up under an undertaking that they would not be harmed, and then killed — with the consequence that the killers and their descendants were called accursed, a curse the Spartans were still invoking politically on the eve of the Peloponnesian War, in 432 BC. He also reports that the Spartans had once removed helot suppliants from the sanctuary of Poseidon at Tainaron and killed them, and that a devastating earthquake was afterwards attributed to the act. Herodotus tells how the people of Kyme, holding the fugitive Pactyes as a suppliant, consulted the oracle at Branchidae, were told to surrender him, and were challenged by Aristodikos, who provoked the god by driving nesting birds from the temple; and how Kleomenes of Sparta burned Argives who had taken refuge in a grove sacred to the hero Argos.

The institution remained live long enough to be audited by an imperial power. Tacitus reports that in AD 22 the Roman senate reviewed the asylum claims advanced by Greek cities and sanctuaries, heard their embassies produce ancient titles and myths in support, and pruned the list. That episode attests three things at once: that the claims were numerous and financially consequential, that the cities defended them with founding legends rather than documents, and that by the Principate the status existed at Rome's discretion. In Ptolemaic Egypt, temples received formally granted rights of asylum recorded on stelai, a parallel case in which a state created, bounded and licensed sanctuary refuge rather than merely acknowledging it.

How the rite worked

Supplication was performed, not declared. The suppliant made physical contact with a person's knees or chin, or took up position at an altar, hearth or cult statue, and often carried a hiketeria — a branch bound with wool — which could be laid on an altar to convert a private appeal into a formal item of public business. The effect of the posture was to shift the burden: the suppliant now belonged, provisionally, to the god, and the community had to decide what to do in the open. In the cases the historians chose to record, direct violence against a suppliant in contact with sacred ground is presented as incurring pollution, agos, which could attach to a family and be invoked against its descendants generations later. Whether that reflects a norm generally observed, or the rhetoric of the particular episodes that got written down, is the second dispute below. The practical consequence was a repertoire of indirect removals — starving suppliants out, luring them off sacred ground with promises, walling them in — precisely the manoeuvres the historians describe with disapproval. F. S. Naiden (Ancient Supplication, 2006) has argued that the essential shape of the act was a petition submitted for evaluation, with a decision that could go against the suppliant, rather than an automatic immunity.

Asylia worked by paperwork and travel. A city decided at home to seek recognition, often anchoring the request in an oracle or a reported epiphany; envoys were appointed and sent, sometimes in teams covering different regions; foreign assemblies debated and passed decrees granting recognition and frequently accepting the associated festival; the texts were carried home and inscribed as a single dossier where visitors would read them. The costs were real — embassies, hospitality, cutting hundreds of lines of stone — and the returns were prestige, festival attendance, and exemption from reprisal seizure. Grants could be sought from kings as well as cities, and a royal grant carried different weight from a civic one.

The physical boundary mattered because the protection was territorial. Sanctuaries were marked out by boundary stones, and the extent of the protected ground at major shrines was a matter of negotiation, extended and cut back by Hellenistic kings and later by Roman authority. The practical limits are visible throughout the record: sanctuaries could be blockaded, suppliants could be starved, and the status conferred by a decree depended on the willingness of armies to observe it. Rigsby's conclusion from the whole corpus was that the grants did not in fact restrain Greek military behaviour.

For particular categories of suppliant there were particular routines. Late lexicographers and commentators report that fugitive slaves used particular shrines, and that what was on offer there was sale to a different owner rather than freedom — an arrangement that would have protected the slave from immediate violence while preserving the institution of slavery. No classical document attests it. Political fugitives and defeated parties in civil conflict appear repeatedly at altars in the historians, and their fates turn on negotiation, guarantees given and broken, and the calculation of how much pollution a community would accept.

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

    Asylia decrees, envoy dossiers and boundary markers document the procedure, geography, chronology and diplomatic scale of inviolability as an institution.

    What it cannot show — A decree records that a city voted to recognise a status; it says nothing about whether anyone subsequently respected it, and no surviving inscription records asylia being successfully invoked to stop a seizure. The dossiers were compiled and displayed by the requesting city, so they are self-selected advertising in which refusals do not appear.

  • Documents

    Ptolemaic temple asylum stelae and related administrative material show a state explicitly granting, delimiting and policing rights of refuge at named temples.

    What it cannot show — This is Ptolemaic royal administration in Egypt, operating on Egyptian temples with their own traditions. It cannot be transferred to the classical Greek polis without argument, and the surviving stelai are clustered in a narrow region and period.

  • Literary sources

    Herodotus, Thucydides, the Attic orators and Tacitus supply the cases: how supplication was performed, how it was evaded, and how violations were exploited politically.

    What it cannot show — Violations are massively over-represented, because a supplication honoured produced no story. Herodotus and Thucydides wrote decades to a century after the events they narrate and shaped them to arguments about pollution, Spartan conduct and Athenian guilt. Tacitus writes as a Roman senator about Greek claims he regards with some irony.

  • Archaeology

    Temene, boundary stones and altar settings define the ground within which protection was claimed and show how it was demarcated.

    What it cannot show — An altar cannot demonstrate that anyone ever took refuge at it, and a boundary stone records where a line was drawn, not whether it held. No excavated feature distinguishes a sanctuary that granted refuge from one that did not.

What the sources do not record

  • The ordinary successful supplication is entirely unrecorded. Every well-known case in the historians is a violation, a crisis or a scandal. Nothing counts the suppliants who were received, sheltered and quietly resettled, which means the surviving evidence systematically misrepresents how the institution normally functioned.

  • No Greek city has left a register of who took refuge in its sanctuaries, on what grounds, or with what outcome. There is no equivalent of a court roll for supplication.

  • We have almost no suppliant's own words. The extended speeches of suppliants that survive are in tragedy, composed by Athenian dramatists for Athenian audiences; the real petitions were oral and are reported, when at all, by others in a few words.

  • Enforcement of asylia is invisible. Among hundreds of recognitions, there is no document recording a case in which the status was invoked against a seizure and prevailed.

  • The position of fugitive slaves at particular refuges is known largely through late lexicographers and commentators rather than contemporary documents, so the rules routinely repeated in modern accounts rest on thin and late testimony.

What the evidence supports

Evidence level: Documented

The scale, procedure and monumental publication of an asylia campaign — a city's diplomatic reach and its willingness to spend heavily on displaying recognition.

The Leukophryena asylia dossier: decrees of recognition from well over a hundred cities, leagues and royal courts, inscribed together with an explanatory text on the sanctuary walls following the festival's first celebration in 208 BC (Magnesia on the Maeander).

Evidence level: Probable

Recognition was sought in coordinated circuits, that mainland Greek bodies and Cretan cities were the target audiences, and that the campaign was entangled with Seleucid royal favour.

The Teian asylia dossier of c. 204/3 BC, including replies from the Aetolians, the Delphians, the Amphictyony, the Athamanians and a series of Cretan cities, obtained by a single set of envoys (Teos, Ionia).

Evidence level: Documented

Inviolability was a widely competed-for civic and religious honour with a documented three-century arc, and that the surviving record is one of grants rather than of enforcement.

The epigraphic corpus of asylia grants assembled in K. J. Rigsby, Asylia: Territorial Inviolability in the Hellenistic World (1996), beginning in the 260s BC and running to the first century BC (Greek mainland, Aegean, Asia Minor and Crete).

Evidence level: Probable

A state explicitly creating and bounding temple refuge by decree — documentary evidence of asylum as an administrative grant rather than an assumed religious given.

Ptolemaic stelai granting and delimiting rights of asylum to named temples (Egypt, principally the Fayum, second to first centuries BC).

Evidence level: Probable

The ground on which protection could be claimed was physically defined and therefore contestable, extendable and reducible.

Boundary markers (horoi) delimiting sacred precincts (Various Greek sanctuaries).

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

    Aeschylus, Suppliants, produced at Athens in the fifth century BC; the same theology appears in Homeric supplication scenes.

    Zeus Hikesios and Zeus Xenios watch over suppliants and strangers, and those who reject or harm them offend the god directly. The Danaids, fleeing marriage, arrive at Argos with suppliant boughs and force the city to choose between war and impiety.

    Offered as explaining — Why the community, and not merely the individual approached, is obliged by a suppliant's claim.

    This is a theological rationale dramatised for a democratic audience deciding matters by vote — the play stages a city debating. It is evidence for how Athenians thought and argued about the obligation, not for a rule of law, and the temptation to reconstruct the institution from these scenes is precisely the error to avoid.

  • Evidence level: Mythological

    Thucydides 1.126–127, writing in the later fifth century about a seventh-century event.

    The killing of Cylon's suppliants left a hereditary curse on the family responsible, an agos that clung to their descendants.

    Offered as explaining — Why a particular Athenian family, the Alcmaeonids, carried an inherited religious taint — and, in 432 BC, why Sparta demanded the expulsion of 'the accursed' from Athens.

    Thucydides presents the story as a live political instrument: the Spartans revived it to embarrass Pericles. It is an explanation attached to a family's status and deployed for advantage, not an independently datable event, and Thucydides himself notes the uses to which it was put.

  • Evidence level: Mythological

    Thucydides 1.128.

    The great earthquake that struck Sparta was the gods' penalty for dragging helot suppliants from the sanctuary of Poseidon at Tainaron and killing them.

    Offered as explaining — A natural disaster, and the Spartans' own subsequent religious anxiety about it.

    A causal claim made by contemporaries about an event, transmitted by a historian who reports the belief. The rite's violation and the earthquake are separately reported; the link between them is the explanation, and it must not be presented as an attested consequence.

Vocabulary

hiketeia
Supplication: the ritualised appeal by which a person placed themselves under divine protection and obliged the individual or community approached to give a public answer.
hiketēs
The suppliant — the person performing that appeal, and thereby, in Greek terms, belonging provisionally to the god.
hiketēria
The suppliant's branch, bound with wool; at Athens, laying one on an altar could formally put a petition before the council or assembly.
asylia
Inviolability: a formally granted status, conferred by decree on a sanctuary, city or individual, exempting them from seizure.
sylē (also rhysia)
The recognised right of reprisal — seizing the persons or goods of citizens of a community that had denied redress. Asylia is the negation of this, and it is the specific hazard the institution addressed.
hiera kai asylos
'Sacred and inviolable' — the standard formula of recognition in Hellenistic asylia decrees.
Zeus Hikesios
Zeus in his aspect as protector of suppliants; the theological warrant to which appeals were addressed.
temenos
The bounded sacred precinct, marked by boundary stones, within which the claim to protection operated.

Contested points

Evidence level: Disputed

Did asylia have practical legal force, or was it essentially an honour?

Rigsby's conclusion from the whole corpus is that, whatever hopes of neutrality lay behind the gesture, the declarations did not in fact alter Greek military behaviour, and that inviolability became primarily a civic and religious honour for which cities competed. Against this, scholars working on Cretan reprisal treaties and on the practice of sylē — Angelos Chaniotis among them — argue that the grants had concrete bilateral effect in exactly the sphere they addressed, exempting the recipient's citizens and territory from a real and frequently exercised right of seizure, and that measuring the institution against the conduct of armies is measuring it against something it never claimed to govern.

Evidence level: Disputed

Was a suppliant at an altar inviolable as a matter of religious rule, or was supplication a petition that could be, and often was, refused?

The older and still common account treats contact with sacred ground as conferring immunity, so that removing or killing a suppliant was a breach of an absolute norm producing pollution — the reading the historians' language of curse and earthquake seems to support. F. S. Naiden (Ancient Supplication, 2006) reconstructs supplication instead as a staged procedure — approach, gesture, plea, and a decision by the person or body supplicated — in which refusal was a normal and legitimate outcome, and argues that the moral outrage in the sources is rhetoric about particular cases rather than evidence of an inviolable rule. The difference decides whether Cylon's suppliants and the Tainaron helots are exceptions or simply the cases someone chose to record.

Evidence level: Disputed

How far can Athenian tragedy be used as evidence for supplication procedure?

One position holds that the suppliant plays — Aeschylus' Suppliants, Euripides' Heraclidae and Suppliants — dramatise real Athenian procedure closely enough to be used as evidence for it, since the audience had to recognise what was happening on stage. The other holds that these plays construct an idealised Athens that receives the rejected and enforces panhellenic norms, a flattering civic self-image staged at a state festival, and that inferring institutional rules from them imports Athenian ideology into the reconstruction of the practice.

Where this comes from

  • Thucydides, History of the Peloponnesian War 1.126–127

    Narrates the Cylonian conspiracy: the survivors take refuge as suppliants at the altar on the Acropolis, are close to death from hunger, are brought up on an undertaking of safety and then killed; the perpetrators and their descendants are thereafter called accursed, and the Spartans later revive the charge for political advantage against Pericles.

  • Thucydides, History of the Peloponnesian War 1.128

    Reports that the Spartans had once raised helot suppliants from the sanctuary of Poseidon at Tainaron, led them away and killed them, and that the violent earthquake that afterwards struck Lakonia was attributed to that act.

  • Thucydides, History of the Peloponnesian War 1.134

    Describes the end of the regent Pausanias at Sparta: he takes refuge in a building attached to the sanctuary of Athena of the Bronze House, is walled in and starved, and is carried out at the point of death so that he should not die on sacred ground.

  • Herodotus, Histories 1.157–160

    Tells how the people of Kyme sheltered the fugitive Pactyes as a suppliant, consulted the oracle at Branchidae and were twice told to surrender him, and how Aristodikos, disbelieving the answer, drove nesting birds from the temple to make the point that the god was demanding of them what he forbade in his own precinct.

  • Herodotus, Histories 6.79–80

    Reports that after the Argive defeat at Sepeia, Kleomenes of Sparta lured some Argive survivors out of the sacred grove by name and killed them, then had brushwood heaped around the grove and burned it, learning only afterwards that it was sacred to the hero Argos.

  • Tacitus, Annals 3.60–63

    Records the Roman senate's review in AD 22 of the asylum rights claimed by Greek sanctuaries and cities, the embassies that came to defend them with ancient legends and royal grants, and the senate's decision to restrict the claims.

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.