THE GETTY HEXAMETERS AND GREEK TRAGEDY

Abstract

The present article aims to show that several passages of Greek tragedy make use of language present in the Getty Hexameters, especially in contexts where incantations and protection of the city are mentioned. The Getty Hexameters were written on a lead tablet at the end of the fifth century BC in Sicily (Selinus or, more likely, Himera). The article argues that the composition of the text predates the lead tablet by several decades (section 2). It focuses on similarities in structure and language that involve Soph. fr. 535 (section 4), Aeschylus’ Oresteia (section 5), Sophocles’ Oedipus Tyrannus (section 6) and Euripides’ Hecuba (section 8). It also suggests that Plato (section 7) and late antique poetry and prose (section 9) reuse some of the linguistic elements of the incantatory tradition of the Getty Hexameters.

1. Introduction

A passage of Timaeus of Tauromenium (fourth/third century BC) discusses the importance of the myth and cults of Demeter and Persephone in Sicily. He claims that Carcinus, a fourth-century BC tragic poet (from Athens or possibly from Akragas),¹ knew the cults of Demeter and Persephone as practised in Syracuse, and was influenced by them in his poetry. This is the fragment of Carcinus quoted by Timaeus:²

λέγουσι Δήμητρός ποτ' ἄρρητον κόρην

Πλούτωνα κρυφίοις ἁρπάσαι βουλεύμασιν

δῦναί τε γαίας εἰς μελαμφαεῖς μυχούς⋅

Demeter's daughter, her whom none may name,

By secret schemings Pluton, men say, stole,

And then he dropped into earth's depths, whose light

Is darkness (tr. Oldfather (1939))

The underworld is described by the striking oxymoronic phrase γαίας … μελαμφαεῖς μυχούς ‘earth's depths, whose light is darkness’. This could be seen as an original image created by an inventive poet. In 2011, David Jordan and Roy Kotansky published the text of a lead tablet from Sicily, dated to the fifth or fourth century BC, which uses the same oxymoronic image to describe the underworld. The text is generally known as the Getty Hexameters, from the name of the museum that houses the artefact. The Getty Hexameters mention Persephone and Demeter, and call the underworld a ‘place whose light is darkness’ (GH (= Getty Hexameters) 8–10):

8  †ͱοσσα† κατὰ σκιερῶν ὀρέων μελαναυγέϊ χώρωι

9  Φερσεφόνης ἐγ κήπου ἄγει πρὸς ἀμολγὸν ἀνάγκη[ι]

10  τὴν τετραβήμονα παῖς ͱαγίην Δήμητρος {ͱ}ὀπηδόν,

†…† the child, from the shadowy mountains in the black-lit place,

Leads from Persephone's garden by force to its milking

The four-footed attendant of holy Demeter…³

The similarities with Carcinus in context and phrasing are striking. The text, of Sicilian origin, is linked to incantations and, according to many interpreters, mystery cults. Carcinus slightly adapts the language because of the different metrical context (μελαμφαεῖς is much easier to insert in iambs than μελαναυγής, suitable for dactyls). Carcinus’ text itself, by using the word ἄρρητον ‘unspeakable’, ‘secret’ in line 1, alludes to the taboo on Persephone's name: her name cannot be revealed to people who are not initiated into the mysteries (Pausanias 8.37.9). The adjective ‘unspeakable’ thus functions as an allusion to mysteries.⁴ Several other tragic texts had already used this phrase (Eur. Hel. 518–19 μελαμφαὲς … ἔρεβος, Hec. 152 νασμῶι μελαναυγεῖ, Alc. 261–2 ὑπ' ὀφρύσι κυαναυγέσι | βλέπων πτερωτὸς Ἅιδας, and below, section 8). Only the similarity between the Getty Hexameters and one of these texts (Eur. Hec. 152 νασμῶι μελαναυγεῖ) has been discussed so far, and from a very different perspective. Some scholars suggest that the Getty Hexameters imitate Soph. fr. 353.2 TrGF and Eur. Hec. 152.⁵ Others argue that both tragedy, in these two cases, and the Getty Hexameters simply use traditional poetic language.⁶

This article aims to show that several other passages of Greek tragedy make use of language present in the Getty Hexameters, especially in contexts where incantations and protection of the city are mentioned. Tragedy incorporates and readapts many elements taken from earlier genres of poetry, from epic to lyric; the sung and spoken performances of choruses and characters often allude to these genres within the tragic setting.⁷ Tragedy also often alludes to magical texts and performances.⁸ The publication of the Getty Hexameters allows modern interpreters to perceive the background of many tragic passages and the allusion to magical practices that would otherwise have eluded attention.

The lead tablet on which Getty Hexameters are inscribed was donated to the Getty Museum in 1981 together with the lex sacra of Selinus, also on lead, and with ‘three curse tablets of the early 5^th c. BCE’, also from Sicily.⁹ Several complete editions of the Getty Hexameters have been published,¹⁰ as well as studies on specific passages.¹¹

Preparing a new full diplomatic and philological edition of the text, including a critical apparatus and palaeographic notes, is beyond the scope of this paper. The line numbers used (from 1 to 50) are those provided in the edition printed in Faraone and Obbink (2013b).¹² The text of the relevant sections will be provided, when necessary, for the convenience of the reader. The translation, unless otherwise noted, is that of Janko (2015), occasionally slightly adapted.¹³ Many points of text and interpretation are uncertain. For reasons of space, textual and interpretive problems will be discussed only when they affect the point discussed in this paper.

Section 2 will illustrate in brief the debate on the date and origin of the Getty Hexameters, arguing that the text was written in Sicily (Selinus or, more likely, Himera), and that the composition of the text predates the artefact (end of the fifth century BC) by several decades. Linguistic and epigraphic data suggest that it is highly unlikely that late fifth-century tragedy influenced the text of the Getty Hexameters as a whole. The main argument of the article (sections 3–9) does not depend on conclusion of this (more technical) section. Sections 3–9 can be read independently from section 2.

Section 3 offers a brief overview of the content of the Getty Hexameters.

Sections 4–6 and 8 discuss the interaction between the Getty Hexameters and Greek tragedy, focusing on similarities in structure and language that involve Soph. fr. 535 (section 4), Aeschylus’ Oresteia (section 5), Sophocles’ Oedipus Tyrannus (section 6) and Euripides’ Hecuba (section 8). Plato (section 7) and late antique poetry and prose (section 9) reuse some of the linguistic elements of the incantatory tradition that reverberate in tragedy.

2. Context, date and language

2.1 Texts and copies

The Getty Hexameters are part of a constellation of texts. Several sections of the Getty Hexameters are transmitted, in identical or similar form, alone or in combination with other texts, in a number of other documents, spanning from the early fifth century BC to the fourth century AD, and coming from different part of the ancient world (Sicily, Crete, Egypt, Rome).¹⁴

Richard Janko attempts to reconstruct the ‘archetype’ of all the extant texts, starting from the Getty Hexameters, the longest text.¹⁵ It is however possible the Getty Hexameters (or their model) were written assembling pre-existing different shorter texts. This longer version is not necessarily the archetype of the shorter text.¹⁶ The shorter texts may have different linguistic and dialectal characteristics from that of the Getty Hexameters. Moreover, it is clear that the text was heavily changed in some copies: lines 36–41 of the Getty Hexameters are written as prose, but clearly derive from a hexametrical text.¹⁷ The present article will focus on the Getty copy as a specific version of a widespread group of similar texts.

2.2 Writing, the date of the tablet and the date of the text

The first editors, Jordan and Kotansky, suggest that the tablet was written in Sicily in the period 425–375 BC.¹⁸ They argue that it could have come from Selinus, like many other similar inscriptions on lead. However, they point out that the use of the ‘half eta’ symbol ͱ to indicate aspiration has a parallel in Himera, not Selinus.¹⁹

Jan Bremmer and others suggested 409 (the date of the destruction of Selinus by Carthage) as a terminus ante quem for the Getty tablet.²⁰ We have indications that Selinus was inhabited after 409, even if many limitations to civic life were in place.²¹ The terminus ante quem of 409 is not relevant if Himera is the place of origin of the text or of the tablet.

The Getty tablet contains, as many scholars have noted, a series of mistakes that originate in writing, not in composition or oral transmission. The text itself stresses the importance of writing and copying. Right at beginning, it presents itself as ‘the meaning-filled letters of these sacred verses … inscribed on lead’ (τῶνδ’ ͱιερῶν ἐπέων ἀρίσημα … | γράμματα κασσιτέρωι κεκολαμμένα, 2–3),²² and thus points to writing as the medium of transmission.

There are several mistakes in the writing of aspirations.²³ Different explanations have been put forward, which require a complex series of adaptation between different linguistic contexts and/or writing systems.²⁴ Two mistakes clearly suggest that the Getty tablet derives at least in part from a written transmission. The spelling ιαγίην in line 10 is very likely to derive from ͱαγίην: the half eta, indicating aspiration, was mistaken for an iota.²⁵ Olga Tribulato plausibly argues that the spelling ͱέπε’ (i.e. ἔπε(α) ‘words’) in line 7 is a copying mistake originating from a written text that read Ϝέπε’. She notes that the digamma is not used in texts from fifth-century Himera or Selinus.²⁶ It is however used in the late sixth-century lex sacra from Selinus and is attested until the mid-fifth century in that city.²⁷ This would be consistent with the hypothesis that the Getty Hexameters derive from a text written many decades before the end of the fifth century, at a time when digamma was in use; the sign was unfamiliar to a late fifth-century scribe and was wrongly transcribed.²⁸ We would thus need to posit an ‘original’ Ionic-Epic text without digamma; we must suppose that it was then transcribed in an area where the digamma was written (at least on some occasions) and the corresponding sound pronounced; and finally, we must suppose yet another transcription which mistakes the digamma (by now out of fashion, or not in use in the area) for an aspiration mark. This projects the text into a past that is very distant from the dating of the Getty tablet.

It seems likely that this text was in circulation, in some form, in the late sixth or early fifth century. This has important consequences not just for the dating of the composition but also for the relationship of the Getty Hexameters to tragedy.

2.3 Language

The Getty Hexameters use a mainly Ionic-Epic language, but several non-Ionic forms are also present. The ‘Doric’ forms include νιν (4),²⁹ σκιαρῶν (8), ἀκαμαντορόα (11), φρασίν (39). We also find instances of Atticism (13 [Ε]ἰνοδία{ι} δ’ ⟨ͱ⟩Εκάτɛ̄{ι}) and words that could be ‘Doric’ or Attic (34 βίαι, dative).³⁰

Tribulato convincingly suggests ‘that the original text employed an Ionic diction mixed with certain metrically guaranteed features adopted from Doric and perhaps consciously employed to gesture towards the linguistic context in which the text was produced’, and connects this with Himera, a city where, according to Thucydides (6.5.1), a dialect ‘between that of the Chalcidians and Doric’ (φωνὴ μὲν μεταξὺ τῆς τε Χαλκιδέων καὶ Δωρίδος ἐκράθη) was spoken.³¹ Many scholars present the provenance from Selinus as a fact, but Himera, as Tribulato argued, is at least as likely as Selinus, if not more.³²

There is another passage in the Getty Hexameters which may suggest the mixing of different linguistic codes. At line 16 the tablet presents the word προμολεισα, corrected into προμολοῦσα in the editio princeps, a correction accepted by all subsequent editors.³³ This is a very strange mistake, since -οῦσα is a very common participle ending, correctly transcribed elsewhere in the tablet (12 πεπιθοῦσα, 14 ἐκκλάζουσα), and the error is unexplainable from a palaeographic or phonetic point of view. One could explain it more easily if the original had the form προμολοῖσα. This Aeolic/Pindaric participle form is common in lyric but unusual in other genres, and it could have prompted the scribe to assimilate the ending to that of participles in -εισα. Forms in -oισα, outside Aeolic poetry, occur in Alcman (e.g. PMG 1.61 φεροίσαις, 3.64 Ἀ[σ]τυμέλοισα and 65 ἔχοισα),³⁴ in the ‘quasi-hexametric’ poetry of Stesichorus of Himera (see e.g. frr. 19.19 ἐχοίσαι, 118.9 δ]ρακοῖσα, 211.17 ].λφεοῖσα in Finglass’ edition),³⁵ in Pindar and Bacchylides,³⁶ in kitharodic poetry,³⁷ in Eumelus (PMG 696.1 Μοῖσα and 2 ἔχοισα (Dindorf, on the basis of Μοῖσα: ἔχουσα MSS)), in Epicharmus fr. 80 (μοισικὰν ἔχοισα: the forms in -οισα are attested as variant readings and are probably the form that Epicharmus used here),³⁸ Rhinthon (fr. 6 Kassel-Austin ἔχοισα). The occurrence of these forms in non-Aeolic poetry are to be explained as borrowing from the Aeolic tradition.³⁹ The forms also occur in hexametric or elegiac epigraphical texts from many areas, including Attica (PMG 938e Μοῖσά μοι), Corinth,⁴⁰ Boeotia,⁴¹ Crete⁴² and Southern Italy.⁴³ These instances in hexametric or elegiac poetry are probably to be interpreted as influenced by the language of lyric poetry. Form in -oισα are of course regularly found in the Theocritean corpus (e.g. 20.39 μολοῖσα), where they may have been felt as part of the ‘Doric’ (Pindaric) elements.⁴⁴ Greek poetical languages regularly insert elements from different traditions into whatever literary language they adopt as primary.⁴⁵ This does not mean that the model of the Greek Hexameters was originally written (if an ‘original’ model ever existed) with the form in -οισα or that we should expect these forms to occur consistently in the poem. It simply means that at some point in time one occurrence of the participle was given this unique non-standard feature, which could have been perceived as a prestigious ‘Lyric’/‘Doric’/‘Stesichorean’ element. That unique form was then corrupted to προμολεισα.

2.4 Conclusions on date and place of origin

On the basis of these palaeographic and linguistic considerations, it is very likely that the tablet itself was written in Sicily (possibly in Himera or Selinus) at the end of the fifth or beginning of the fourth century BC. The text itself (or at least large sections of it) must have been composed some decades before it was inscribed on the Getty lead tablet; it was transmitted in writing, probably in different Greek-speaking contexts and using different alphabetic systems.

3. Structure of the text

The Getty Hexameters are structured according to a very clear pattern. We find four occurrences (with slight variations) of a line invoking Paean. These invocations divide the text into clearly marked sections which differ for content and style. This is the outline:

• 1–5   the speaker asserts the efficacious power of the incantation

• 6  invocation to Paean:

  Παιήων, σὺ δὲ πάντοσ’ ἀλέξιμα φάρμακα πέμπεις
  ‘Paeon – to every place you send protective drugs – ’

• 7–22   Paean narrates a story about a goad, related to Persephone and Hecate

• 23  invocation to Paean:

  ⸤Παιήων,⸥ σὺ γὰρ αὐτὸς {ͱ}ἀ⸤λ⸥έξιμα φάρμα⸤κα πέμπεις.⸥
  ‘[Paeon] – you send protective drugs yourself – '

• 24–31  the text states that incantations will protect ships, people, animals and the city in general

• 32  invocation to Paean:

  [Παιήων, σὺ δ]ὲ πάντοσ’ ἀκεσ{σ}φόρος ἐσσὶ καὶ ἐσθ[λός.]
  ‘[Paeon] – to every place you bring cures and are good.’

• 33–48  Ephesia grammata

• 49  invocation to Paean:

  [Παι]ήων, ͱο γὰρ αὐτὸς ἀλέξιμα φάρμακα πέ[μπει⋅]
  ‘[Paeon] – since he sends send protective drugs himself –’

• 50  conclusion: Paean will protect against evil pharmaka.

Refrains are of course common in magical texts (e.g. Aesch. Eum. 328–32 = 341–6; Theocritus 2.17), but also in paeans.⁴⁶

Interpretations of the Getty Hexameters vary. Jordan and Kotanksy consider them the ‘traditional legomena of a rite of initiation into the worship of Demeter and Core’.⁴⁷ Similarly, Dirk Obbink thinks that this text is ‘poetry performed in a mystery context’ and calls it a ‘telestic song’.⁴⁸ William Furley argues that ‘the Getty narrative is nothing other than the sacred narrative of Dionysos’ birth’, and that its author is ‘an Orpheotelestēs’ who ‘appeals to Apollo Paian as the divine healer, who, according to myth, had even saved Dionysos as a child’;⁴⁹ the first line ‘introduces the hieros logos of mystery rites’.⁵⁰ Janko instead considers it ‘a late fifth-century hexametric incantation against witchcraft, into which an earlier spell is embedded; it was intended for civic rather than private use’;⁵¹ Christopher Faraone and Radcliffe Edmonds also argue that the text is a civic incantation.⁵²

Explicit references to mystery cult are not present in the extant part of the text, and the references to Persephone (8) and to the Idean Dactyl Damnameneus (41) do not necessarily prove that this was used in a mystery context. The emphasis on the protection of the polis (25–31) indicates that this a text was part of a ritual aimed not at a single individual but at the community as a whole.⁵³

Ian Rutherford, focusing on the paeanic features, suggests that the Getty Hexameters can be ‘a paean-incantation, in fact the only example of this sub-genre to survive’.⁵⁴ One should add that the dactylic rhythm is very frequent in paeans.⁵⁵ Richard Janko and Carlo Martino Lucarini accept that this text is a paean in hexameters, comparable to the one composed by Socrates before his death.⁵⁶ The hypothesis that this text was felt to be a Paean-incantation is compatible with its status as amulet: the first three lines of the text allude both to the oral performance of the text (GH 1 ἐπαείδω ‘I sing’) and to the fact that the text is written down on lead (GH 2–3 τῶνδ’ ͱιερῶν ἐπέων ἀρίσημα … | γράμματα κασσιτέρωι κεκολαμμένα ‘the meaning-filled letters graved on lead’).

The structure and language of the Getty Hexameters is reflected in some passages of Greek tragedy, especially Aeschylus. This suggests that similar Paean-incantations were more widely used and known. Aeschylus, Sophocles and Euripides all use the language of magic and incantation as preserved in the Getty Hexameters. It is certainly possible in principle that the influence went both ways, and that magical texts occasionally imitated specific tragic phrases and epithets. The interaction of genres is however in the other direction: tragedy clearly alludes to the genre of magic/religious poetry in many passages;⁵⁷ magic texts do not need an allusion to the tragic genre for their literary and performative goals.

In what follows, to indicate an allusion to a genre or a sub-genre such as ‘incantations’ or ‘paean’ (as opposed to the instantiation of the genre in a specific text), we will speak of ‘genre intimation’. To indicate the use of language that evokes a specific genre, we will talk of ‘linguistic intimation’. To indicate an allusion to a specific text, identified by allusion to its wording, we will speak of ‘textual allusion’; the text alluded to will be called the ‘exemplar-model’ (‘modello-esemplare’ in Conte's influential terminology); ‘intertextuality’ will be used as a general term indicating all different kinds of relationship between texts (allusion, intimation, quotation, similarity).⁵⁸

4. Sophocles fr. 535 (Rhizotomoi)

Bremmer notes that lines 12–13

… [Ε]ἰνοδία{ι} δ’ ⟨ͱ⟩Εκάτɛ̄{ι} φρικώδεϊ φωνῆι

⸤βά⸥ρβαρον ἐκκλάζουσα θεὰ θεῶι ͱηγεμονεύ⸤ει⋅⸥

Hecate of the crossroads, screaming obscurely

In hair-raising voice, a goddess, leads the god

present similarities with Sophocles’ fr. 535, a choral passage from the play Rhizotomoi ‘Root-cutters’.⁵⁹ The play included a description of Medea's magic abilities.⁶⁰ The fragment runs as follows:

Ἥλιε δέσποτα καὶ πῦρ ἱερόν,

τῆς Εἰνοδίας Ἑκάτης ἔγχος,

τὸ δι' Οὐλύμπου <προ>πολοῦσα φέρει

καὶ γῆς ἀνιοῦσ’ ἱερὰς τριόδους,

στεφανωσαμένη δρυῒ καὶ πλεκταῖς

ὠμῶν σπείραισι δρακόντων

O Sun our lord and sacred fire, the spear of Hecate of the roads, which she carries as she attends her mistress in the sky and as she goes up the sacred crossroads of the earth, crowned with oak-leaves and the woven coils of savage dragons!⁶¹

The similarities consist in the phrase ‘Hecate of the roads’, in the mention of torches (‘spear’) and the procession. It is unlikely that the widely attested epithet of the goddess was invented by Sophocles in this fragment.⁶² Sophocles is likely to have imitated cultic language (‘linguistic intimation’); he clearly introduces an innovative metaphor (‘spear’) to describe a cultic object (‘torch’), but the metaphor is not present in the Getty Hexameters. Many texts connect torches with Hecate and mention her role as ‘attendant’ and the procession.⁶³ These considerations do not prove that the Getty Hexameters imitate the passage of Sophocles. Nor could one argue in the opposite direction. As one anonymous reader put it, ‘in a world where it is entirely plausible that there were processions in connection with Hecate of the crossroads, no link at a textual level is required’. It is legitimate to see ‘linguistic intimation’ to cultic language in Sophocles, but not a ‘textual allusion’ by Sophocles to the Getty Hexameters or vice versa.

5. The Oresteia of Aeschylus

The Oresteia provides some striking parallels for the language and genre of the Getty Hexameters (‘linguistic intimation’ of magic language in tragedy). Incantations and magic are often discussed in the trilogy, at crucial moments: in the Agamemnon the most striking passages occur in the third stasimon, in Clytemnestra's speech to Agamemnon (958–72), and in Thyestes’ curse, as reported by Aegisthus (1601–2); in the Choephori, in the second stasimon; in the Eumenides, in the incantation of the Furies in the first stasimon.⁶⁴ This section will show that the connection between the Oresteia and magic texts, in particular the Getty Hexameters, is stronger than simple linguistic intimation.

5.1 The Agamemnon

Let us start with the parodos of the Agamemnon. The chorus, just like the speaker of the Getty Hexameters, proclaims their authority (GH 1 οὐκ ἀτέλεστ’ ἐπαείδω ‘and the spells that I sing are not unfulfilled’; Aesch. Ag. 104–5 κύριός εἰμι θροεῖν ὅδιον κράτος αἴσιον ἀνδρῶν ἐντελέων⋅ ‘I have authority to tell of the auspicious departure of the commanders, | men invested with power–’).⁶⁵ The chorus here speaks in hexameters, suggesting a link with the epic and kitharodic tradition;⁶⁶ the Getty Hexameters form part of this tradition too.

There are similarities in content. The Getty Hexameters envisage a crisis situation: ‘a doom that brings death’ (26–7) threatens humans and flocks; the crisis can be meet by repeating ‘night and day’ (29) a prayer or incantation, saying ‘[better] so for the city’ (31); the sequence ends with an invocation to Paean at 32. This is the text (23–32):

⸤Παιήων,⸥ – σὺ γὰρ αὐτὸς {ͱ}ἀ⸤λ⸥έξιμα φάρμα⸤κα πέμπεις.⸥ –

κηληθ]μοῦ κατάκουε φ[ρ]ασὶν γλυκὺν ͱ[ύμνον ἐύφρων⋅

πᾶσιν δ’ ἀ]νθρώποισιν ἐπιφθέγγεσ{σ}θαι ἄν[ωγα,       25

ͱως δήμ]ωι κἀν εὐπολέμωι καὶ ναυσίν, ͱότα[ν κὴρ]

ἄφνω ἐπ’ ἀ]νθρώποις θανατηφόρος ἐγγύ[θεν ἔλθηι]

ͱως καὶ ἐπ]ὶ προβάτοις καὶ ἐπὶ τέχναισι βροτ[είαις.

ͱούτω δὴ φ]θέγγε{σ}σθαι ἐν ἐϋφρόνηι ἠδὲ κατ’ [ἦμαρ,

χρησμὸ]ν ἔχων ͱόσιον {σιον} στόματος θυ[ρέτροισιν ἐν αὐτοῖς⋅    30

“βέλτιον ἔ]σ{σ}τι πόλει⋅ τὰ γὰρ ἀρχῆς ἐστιν ἄριστα”.

Παιήων, σὺ δ]ὲ παντόσ’ ἀκεσ{σ}φόρος ἐσσὶ καὶ ἐσθ[λός.

[Paeon] – you send protecting drugs yourself –

[Kindly] hear in your mind the incantation's sweet song.

[I bid you] intone [it] for [all of the] people,     25

As in [folk] good at war and in ships, when [some doom]

[Comes suddenly] nigh bringing men to their deaths,

[As too both] for flocks and for mortal men's crafts,

Utter [what follows] by night and by day,

Keeping holy your [oracle in] the doors of your mouth:     30

‘[Better] so for the city: for order is best.’

[Paeon] – to every place you bring cures and are good.

Many supplements are uncertain, but the general sense seem to be well captured by Janko's text as presented above. An important point for the following discussion is the text of GH 31. Different supplements can be proposed. Kotansky suggests [χρήσιμα γ’ ἔ]στι πόλει and Furley λώιον ἔ]στι πόλε.⁶⁷ One can think e.g. of [βέλτιστ’ ἔ]σστι πόλει, [κάλλιστ’ ἔ]σστι πόλει, [εὖ τάδε γ’ ἔ]σστι πόλει, [εὖ νυ τάδ’ ἔ]σστι πόλει, [εὖ πάντ’ ἔ]σστι πόλει. Janko is in any case probably right to suppose that GH 31 is a direct speech, which must be ‘spoken night and day’ (29) by human beings who must pronounce or keep something ‘holy’ in their ‘mouth’ (30). Janko provides some modern comparisons for this type of repeated statement,⁶⁸ but we will see that the Oresteia offers ancient parallels.

In the parodos of the Agamemnon we find again a crisis that threatens the survival of the community (in this case, the army) that is at war (cf. GH 26 δήμ]ωι κἀν εὐπολέμωι: the text is uncertain but the mention of war is not). The army suffers from hunger (Ag. 188, 194), and the winds destroy ships and cables (Ag. 195 ναῶν <τε> καὶ πεισμάτων ἀφειδεῖς; cf. GH 26 ναυσίν). This crisis is announced earlier in the parodos, when the chorus mentions the omen (the pregnant hare killed by the two eagles: Ag. 109–20) and intones the famous refrain (Ag. 121 = 139 = 159):

αἵλινον αἵλινον εἰπέ, τὸ δ' εὖ νικάτω

Cry sorrow, sorrow, but may good prevail!⁶⁹

This refrain again recalls the hexameter (five dactyls, with the first eight syllables corresponding to the first part of the hexameter until the ‘feminine’ caesura in the third foot). The refrain thus frames, as in the Getty Hexameters, narrative sections that contain both the description of evils threatening a community, and possible remedies. The text also recalls GH 31 [βέλτιον ἔ]σ{σ}τι πόλει ‘[better] so for the city’. In fact, in the first section the prophet Calchas explains that ‘destiny’ is threatening the Trojan community and its animals (Ag. 128–30 πάντα δὲ πύργων | κτήνη πρόσθε τὰ δημιοπληθέα | Μοῖρα λαπάξει πρὸς τὸ βίαιον ‘in front of their walls | Destiny will violently plunder | all the mass of livestock the community possesses’). The loss of livestock (Ag. 129 κτήνη; cf. GH 28 προβάτοις) is of course disastrous for ancient societies, and we will see it mentioned again in the Oedipus Tyrannus (see below, section 6), in a context that also recalls the invocation to Paean of the Getty Hexameters. The prophet explains that the expedition on Troy is at risk because of the wrath of Artemis, a threat of famine and destruction to the Greek community (Ag. 134–7, 194). After the second occurrence of the refrain (139), the speech of the prophet resumes. Note that the Getty Hexameters too are structured as one or possibly two direct speeches of Paean, interrupted by addresses to Paean himself (above, section 3). The refrain ‘may good prevail!’ is a protective charm (repeated three times) for the Greek army, and the prophet stresses that the divinity that might help is Paean himself (Ag. 147–51), again in a passage that recalls the hexameter (seven dactyls in lines 148–9 and 150–1):⁷⁰

ἰήιον δὴ καλέω Παιᾶνα     147

μή τινας ἀντιπνόους Δαναοῖς χρονίας ἐχενῇδας ἀπλοίας  148–9

τεύξῃ, σπευδομένα θυσίαν ἑτέραν, ἄνομόν τιν', ἄδαιτον  150–1

So I call on the healer Paean:

Let her not cause any persistent adverse winds that hold back the Danaan ships

From sailing, bent on another sacrificial slaughter, one without music of feasting.

The phrase of line 147 follows a pattern that we find in lines C–D of the Phalasarna tablet (one of the magical texts that partly overlap with the Getty Hexameters):⁷¹

Ζῆνα τ’ ἀλεξίκακον καὶ Ἡρακλέα πτολίπορθον,

Ἰατρὸν καλέω καὶ Νίκην καὶ Ἀπόλλω[να]

I call on Zeus the averter of ills, Herakles the sacker of cities, Iatros, Nike, Apollon.

The pattern ‘I call on X’ is often attested as a formula of invocation in classical Greek literature.⁷² It clearly recalls the cultic setting that we find in the Phalasarna tablet.

Calchas echoes incantatory and magical language (‘linguistic intimation’); but his invocation fails to avert disaster completely. The Danaan ships do manage to sail, after all, but only at the cost of an unholy/unmusical sacrifice (θυσίαν ἑτέραν, ἄνομόν τιν', ἄδαιτον).

Both the Getty Hexameters and the prophet in the Agamemnon call on Paean to save the community from a crisis; and in both cases the crisis involves Hecate/Artemis. The goddess that is threatening the Greek community in the Agamemnon is Artemis and she is identified with Hecate in a probable interpretation. Alan Sommerstein and Enrico Medda print the text transmitted by the manuscripts TF (Ag. 140–1):⁷³

τόσον περ εὔφρων ἁ καλὰ

δρόσοις ἀέπτοις μαλερῶν λεόντων

So very kindly disposed is the Fair One

To the unfledged seed of fiery lions.

ἁ καλὰ TF: καλὰ MV: Ἑκάτα Badham

Martin West prints Charles Badham's bold (and attractive) conjecture ‘so very kindly disposed is Hecate’.⁷⁴ Even scholars who do not print the correction accept that plausibility of the connection between Artemis and Hecate here.⁷⁵ Aeschylus (Supp. 677) and Euripides (Pho. 109–10) gave Artemis the epithet Hecate.⁷⁶ Already Stesichorus identified Iphigenia with Hecate (F 178 Finglass). The Catalogue of Women from the Hesiodic corpus calls Agamemnon's daughter ‘Iphimede’ and states that after her sacrifice she was made immortal and transformed into Artemis ‘of the roads’ (an epithet that is typical of Hecate: see above, section 4).⁷⁷ The connections between Hecate and the sacrifice of Iphigenia are very strong.

In the Agamemnon we thus find Hecate and Paean, as well as a refrain that imitates the auspice of GH 31 ‘[better] so for the city’: the chorus of the Agamemnon, in the context of Calchas’ invocation to Paean, express their hope that ‘good may prevail’ (τὸ δ' εὖ νικάτω). The refrain is taken up again by Agamemnon himself later in the parodos. The monologue in which he expresses his decision to sacrifice his daughter ends with the sentence (217) εὖ γὰρ εἴη ‘May all be well.’ Agamemnon echoes the refrain of the chorus, but the wish that ‘good may prevail’ is shown to be obviously void at this moment of crisis (202–4). This echo retrospectively calls into question the efficacy of Calchas’ initial appeal to Paean. The chorus voice their disapproval of Agamemnon's decision in the strongest possible terms (218–27). Not only that: the sacrifice of Iphigenia substitutes and represses the Paean that the prophet invoked at 147. Agamemnon, in a situation that inverts the pattern of purity of utterance described in GH 30 (χρησμὸ]ν ἔχων ͱόσιον {σιον} στόματος θυ[ρέτροισιν ἐν αὐτοῖς ‘keeping holy your [oracle in] the doors of your mouth’) and the invocation to Paean in GH 32 (Παιήων, σὺ δ]ὲ παντόσ’ ἀκεσ{σ}φόρος ἐσσὶ καὶ ἐσθ[λός ‘[Paeon] – to every place you bring cures and are good’), has his helpers gag his daughter, who used to sing a paean in her house. Compare Ag. 235–7:

στόματός τε καλλιπρώιρου

φυλακᾶικατασχεῖν

φθόγγον ἀραῖον οἴκοις.

And by putting a guard

On her fair face and lips to restrain

Speech that might lay a curse on his house.

Iphigenia is threatening to issue a ‘curse’ from her ‘mouth’, the very opposite of the ‘pure’ oracle that the ‘mouth’ mentioned in GH 30 must observe. The chorus then explicitly comment on the fact that Iphigenia used to sing a paean in her father's halls (Ag. 243–7):

ἐπεὶ πολλάκις

πατρὸς κατ' ἀνδρῶνας εὐτραπέζους

ἔμελψεν, ἁγνᾶι δ' ἀταύρωτος αὐδᾶι πατρὸς  245

φίλου τριτόσπονδον εὔποτμον

παιῶνα φίλως ἐτίμα.

Because often

At the rich banquets in her father's dining-chambers

She had sung, a pure virgin with pure voice,

Duly and lovingly performing her father's

paean for good fortune to accompany the third libation.

Iphigenia's suppressed paean thus substitutes Calchas’ invocation to Paean at 147: Paean is evoked by the prophet but not sung and remains inefficacious. The chorus does not sing a paean but a refrain to avert evils – but they strongly condemn as evil the sacrifice. Agamemnon echoes the refrain of the chorus when he decides to sacrifice his daughter. The sacrifice stops Iphigenia from performing an incantation against the house (φθόγγον ἀραῖον οἴκοις). The sacrifice evokes the paean sung by Iphigenia in the past and cancels the possibility of a paean by her. This sacrifice suggests the impossibility of singing a paean and of averting evil in the circumstances.

The Getty Hexameters offer ritual and linguistic parallels: the refrain invoking Paean as an incantation against evil (6, 23, 32, 49), the purity of the tongue (30) and the wish that ‘this is better for the city’ (31; cf. Ag. 121 = 139 = 159 ‘may good prevail’, 217 ‘may all be well’). The parodos is thus structured as an incantation that succeeds and fails at the same time: it manages to solve the crisis caused by adverse weather conditions but fails to achieve the prevalence of good that its incantatory refrain was aiming to achieve. The allusion to rituals for Paean and Hecate will not bring ‘good’ to Agamemnon: his echo of Calchas’ prayer is a self-defeating one.

It is impossible to indicate textual signs that prove a direct allusion in the text of the Agamemnon to a specific passage in the Getty Hexameters (‘textual allusion’). However, the linguistic, structural and cultic similarities (refrains, appeal to Paean, curses, Hecate/Artemis) suggest not simply a ‘genre intimation’ and ‘linguistic intimation’ but also an allusion to a magic text that had a very similar structure to the Getty Hexameters.

5.2 The Choephori

The chorus of the second play of the trilogy take up that wish, in another passage that explicitly mentions magic and the welfare of city. In the second stasimon, the chorus imagine that Hermes will help Orestes kill Aegisthus and Clytemnestra and become king of Argos. This is what the chorus imagine they will do after Orestes’ victory (Cho. 819–26)

καὶ τότ' ἤδη κλυτὸν

δωμάτων λυτήριον    820

θῆλυν οὐριοστάταν

†ὁμοῦ κρεκτὸν† γοήτων νόμον

μεθήσομεν⋅ “πόλει τάδ' εὖ⋅

ἐμὸν ἐμὸν κέρδος αὔξεται τόδ’. Ἄ-

τα δ' ἀποστατεῖ φίλων.”  825

And then at last, for the voyage

That will set the house free,

We will utter feminine strains to set the wind fair, the strains that are sung

By charmers to <shrill> accompaniment: ‘it is good for the city!

My gain, my gain is swelling here,

And destruction stands far from my friends.’ (text and translation Sommerstein (2008), adapted)⁷⁸

Note how close the sentence at 823 πόλει τάδ' εὖ is to GH 31 [βέλτιον ἔ]σ{σ}τι πόλει (or [εὖ πάντ’ ἔ]σστι πόλει or other similar supplements).⁷⁹ The chorus is explicitly saying that the words they are going to pronounce are those of ‘charmers’, ‘magicians’, people who make incantations (γοήτων νόμον).⁸⁰ Their incantation-song sets a fair wind (οὐριοστάταν ‘steady and prosperous’; cf. οὔριος ‘with a fair wind’), i.e. the goal of the invocation to Paean in the parodos of the Agamemnon: the fair wind is produced by means of an incantation, as indeed in Empedocles and other magical rituals.⁸¹ This appears to be a direct echo of the incantation language attested in the Getty Hexameters. We have an explicit reference (not an ‘intimation’ or allusion) to the language and words used by ‘magicians’, ‘charmers’, and, at that very moment, a sentence that is very close in wording and content to a sentence found in the Getty Hexameters. This is as close to a textual allusion as it gets; of course, it is possible that the sentence used in the Getty Hexameters (the ‘exemplar-model’) was also used in other, unknown, magic texts, and that the allusion is not to this specific text.

The chorus of the Choephori contains another striking echo of the language (‘linguistic intimation’) of the Getty Hexameters. During the lament for Agamemnon, the chorus express the following wish (Cho. 368–71):

πάρος δ' οἱ κτανόντες νιν οὕτως δαμῆναι,

<x -> θανατηφόρον αἶσαν

πρόσω τινὰ πυνθάνεσθαι     370

τῶνδε πόνων ἄπειρον.

But rather should his killers have been slain so,

<So that> someone far away

Would have learned of their deadly fate  370

Without experiencing these present troubles …

This echoes the phrase found in GH 26–7 ͱότα[ν κὴρ] | ἄφνω ἐπ’ ἀ]νθρώποις θανατηφόρος ἐγγύ[θεν ἔλθηι] ‘when [some doom] | [comes suddenly] nigh bringing men to their deaths’. κήρ is a virtually certain supplement by Burkert (no other suitable word would fill the lacuna so well).⁸² As we will see (below, sections 6 and 7), the adjective θανατηφόρος is very rare in classical, especially poetic, texts: it occurs in a series of passages that seem to allude to the same context as the Getty Hexameters. Another passage in the Choephori that recalls the language of the Getty Hexameters occurs in the first stasimon. The Getty Hexameters state that, thanks to their incantatory powers (GH 4–5),

οὔ νιμ πημανέουσιν ͱόσα τρέφει εὐρεῖα χθὼν,

οὐδ’⟨ͱ⟩όσα πόντωι βόσκει ἀγάστονος Ἀμφιτρίτη⋅

No creature that the broad earth rears shall cause him harm,

No creature in the sea that roaring Amphitrite feeds at sea.

This of course echoes epic language (e.g. Il. 11.741 ἣ τόσα φάρμακα ἤιδη ὅσα τρέφει εὐρεῖα χθών, Hom. Hymn Ven. 5 ἠμὲν ὅσ' ἤπειρος πολλὰ τρέφει ἠδ' ὅσα πόντος) but is very close to Cho. 585–92:

πολλὰ μὲν γᾶ τρέφει

δεινὰ δειμάτων ἄχη.

πόντιαί τ' ἀγκάλαι

κνωδάλων ἀνταίων

βρύουσι⋅ βλάπτουσι καὶ πεδαίχμιοι

λαμπάδες πεδάοροι,     590

πτανά τε καὶ πεδοβάμονα⋅ κἀνεμόεντ' ἂν

αἰγίδων φράσαις κότον.

The earth breeds many beings

That cause terrible, fearful suffering,

And the bosom of the deep

Teems with hostile monsters;

Torches flaming on high,

Between sky and earth, do injury

To winged and footed creatures, and one might also speak

Of the windy wrath of hurricanes.

Note also the chorus’ mention of ‘torches flaming on high’ (590 λαμπάδες πεδάοροι; cf. GH 13 [λ]αμπάδας, Hecate's torches), and the similarity between 591 πεδοβάμονα ‘footed creatures’ and GH 10 τετραβήμονα (the ‘four-footed’ goat lead by the mysterious ‘child’).

These parallels show the pervasive presence of incantatory language (‘linguistic intimation’) in the Oresteia.

5.3 Incantatory language and the Oresteia

These parallels do not prove that the parodos of Agamemnon alludes to the Getty Hexameters specifically (‘textual allusion’). They however suggest that the Oresteia alludes (‘linguistic intimation’) to some phrases from this magical tradition, and to the ritual practices that accompanied them (cf. Ag. 121 = 139 = 159 τὸ δ' εὖ νικάτω and 217 τὸ δ' εὖ νικάτω with GH 31 [βέλτιον ἔ]σ{σ}τι πόλει, Ag. 147 ἰήιον δὲ καλέω Παιᾶνα with Phalasarna lines C–D Ζῆνα τ’ ἀλεξίκακον καὶ Ἡρακλέα πτολίπορθον, | Ἰατρὸν καλέω, Cho. 369 θανατηφόρον αἶσαν with GH 26–7 [κὴρ] | … θανατηφόρος, Cho. 823 πόλει τάδ' εὖ with GH 31 [βέλτιον ἔ]σ{σ}τι πόλει). Not only that: the parodos of the Agamemnon alludes to a magical/ritual text with paeanic refrains (in the tradition of the Getty Hexameters) (‘genre intimation’). In the Agamemnon, however, this paeanic incantation/prayer fails and is substituted by a sacrifice (in which the paean is suppressed). The Choephori specifically mentions the ‘song of people who make incantations’ (822 γοήτων νόμον) and seems to allude to a phrase of that is present in the Getty Hexameters (possibly a ‘textual allusion’). The tradition of incantation that was found in the Getty Hexameters thus allows us to see the poetic and magic–religious background of some crucial moments in the trilogy, and the actual use in incantatory texts of the language employed by Aeschylus.

6. The Oedipus Tyrannus of Sophocles

Another striking similarity with the Getty Hexameters is found in the Oedipus Tyrannus of Sophocles. The prologue explained that ‘the city […] is grievously tossed by storms’ (22–3 πόλις … ἄγαν | ἤδη σαλεύει) since (OT 25–7)

φθίνουσα μὲν κάλυξιν ἐγκάρποις χθονός,

φθίνουσα δ' ἀγέλαις βουνόμοις, τόκοισί τε

ἀγόνοις γυναικῶν

a blight is on the buds that enclose the fruit, a blight is on the flocks of grazing cattle and on the women giving birth, killing their offspring.

In the parodos, the chorus describe the effects of the blight that affects the city: sickness, agricultural crisis, deaths. We already examined (section 5.1) the passage in the Getty Hexameters that envisages a similar crisis: ‘a doom that brings death’ threatens human beings and flocks (25–8); the city repeats ‘night and day’ (29) a prayer or incantation (31); the sequence ends with an invocation to Paean at GH 32. The chorus, in the second antistrophe of the parodos, call on Paean (OT 179–88):⁸³

Ὧν πόλις ἀνάριθμος ὄλλυται⋅

νηλέα δὲ γένεθλα πρὸς πέδῶι  180

θαναταφόρα κεῖται ἀνοίκτως⋅

ἐν δ' ἄλοχοι πολιαί τ' ἔπι ματέρες

ἀκτὰν παρὰ βώμιον ἄλλοθεν ἄλλαι

λυγρῶν πόνων ἱκετῆρες ἐπιστενάχουσι.  185

Παιὰν δὲ λάμπει στονόεσσά τε γῆρυς ὅμαυλος⋅

ὧν ὕπερ, ὦ χρυσέα θύγατερ Διός,

εὐῶπα πέμψον ἀλκάν.

Countless are their deaths, and the city is perishing, unpitied her children lie on the ground, carried off by death, with none to lament; and by the row of altars wives and white-haired mothers on this side and on that groan as suppliants on account of their sad troubles. Loud rings out the hymn to the Healer and the sound of lamentation with it! For these things, golden daughter of Zeus, send the bright face of protection!

Not only is the situation similar (death threatens humans and animals, and the city as a whole). The chorus speaks of a (180–1) γένεθλα … θαναταφόρα ‘children … carried off by death’: cf. the ‘doom that brings death’ of GH 26–7, κὴρ | … θανατηφόρος. In both texts this is followed by an invocation to Paean (for Soph. OT 186 Παιὰν δὲ λάμπει cf. GH 32 [Παιήων]). As Patrick Finglass points out, the ‘focus on Apollo, coupled with the paeanic refrain (153–5n.) and the explicit reference to the paean (187), suggests that the song should itself be characterised as a paean, a type of song with prominently apotropaic associations (4–5n.)’.⁸⁴ The two texts are also connected by metrical similarities. The parodos is predominantly in dactylic metres: the first strophic pair starts and ends with hexameters,⁸⁵ and the dactylic rhythm is present in the second strophic pair as well, mixed with iambs.⁸⁶ In particular, line 181 θαναταφόρα κεῖται ἀνοίκτως corresponds to the second part of a hexameter after the caesura at the third longum.

The word θανατηφόρος is not very common in classical and archaic Greek.⁸⁷ It occurs in poetry only in the Getty Hexameters and in the two passages of tragedy quoted above (Aesch. Cho. 369; Soph. OT 181), passages which have strong verbal and content similarities to the passage in the Getty Hexameters. It is not surprising that both Aeschylus and Sophocles, when describing a crisis that affects a whole community, both alluded to ritual practices that were in actual use in the Greek world. We have here in the Oedipus Tyrannus a strong ‘linguistic intimation’, accompanied by references to very similar ritual and religious practices.

7. Plato's Republic

Plato seems to echo the language of incantation (‘linguistic intimation’) in his myth of Er. When the souls of the dead emerge from the cycle of purification, they encounter a prophet of Lachesis who proclaims (Resp. 617d6–e3):

Ἀνάγκης θυγατρὸς κόρης Λαχέσεως λόγος. Ψυχαὶ ἐφήμεροι, ἀρχὴ ἄλλης περιόδου θνητοῦ γένους θανατηφόρου. [617e] οὐχ ὑμᾶς δαίμων λήξεται, ἀλλ' ὑμεῖς δαίμονα αἱρήσεσθε. πρῶτος δ' ὁ λαχὼν πρῶτος αἱρείσθω βίον ὧι συνέσται ἐξ ἀνάγκης.

The word of Lachesis, virgin daughter of Necessity! Transient souls, the start of another mortal cycle of the human race! A daimon will not be allotted to you, but you will choose a daimon. The one who draws the first lot is to make the first choice of a life, to which he will be bound by necessity. (tr. Halliwell (1988))

Necessity, as Proclus noted, speaks in a solemn and impressive style (note the absence of articles and verbs).⁸⁸ The similarities in content and language with the Getty Hexameters are notable, not simply for the use of the word θανατηφόρος. Both texts describe the afterlife, and both mention Ananke, ‘Necessity’. The Getty Hexameters speak of Persephone's garden (surely the world after death) and Ananke in the same line: Φερσεφόνης ἐγ κήπου ἄγει πρὸς ἀμολγὸν ἀνάγκη[ι] (‘a child leads out of Persephone's garden by necessity for milking’, 9). Plato obviously knows well the language of incantations and magic.⁸⁹ He here uses words that recall language used in actual magical texts.⁹⁰ Already Plutarch (Quaest. conv. 715c) noticed the connection between this passage of Plato and Empedocles’ mention of Necessity in connection with reincarnation;⁹¹ Empedocles’ Sicilian origin, and his interest in magic, are obvious points of contact with the Getty Hexameters.

8. The Hecuba of Euripides and Carcinus

A passage from the Getty Hexameters presents many similarities with a passage from the Hecuba of Euripides. This is the text of lines 8–14:

†ͱοσσα† κατὰ σκιερῶν ὀρέων μελαναυγέϊ χώρωι

Φερσεφόνης ἐγ κήπου ἄγει πρὸς ἀμολγὸν ἀνάγκη[ι]

τὴν τετραβήμονα παῖς ͱαγίην Δήμητρος {ͱ}ὀπηδόν,    10

αἶγ’ ἀκαμαντορόα νασμοῦ θαλεροῖο γάλακτος

βριθομένη<ν>⋅ ͱέπεται ⟨δὲ⟩ θεαῖς πεπιθοῦσα φαειναῖς

⸤λ⸥αμπάδας⋅ [Ε]ἰνοδία{ι} δ’ <ͱ>Εκάτɛ̄{ι} φρικώδεϊ φωνῆι

⸤βά⸥ρβαρον ἐκκλάζουσα θεὰ θεῶι ͱηγεμονεύ⸤ει⋅⸥

†…† the child, from the shadowy mountains in the black-lit place,

Leads from Persephone's garden by force to its milking

The four-footed attendant of holy Demeter,     10

The goat that weighs heavy with rich milk's tireless flow.

She follows, obeying goddesses with torches ablaze.

Hecate of the crossroads, screaming obscurely

In hair-raising voice, a goddess, leads the god.

Bremmer focused on the similarity with Hec. 151–2, but in fact the whole sequence starting at 141 is of importance (Hec. 141–52):

ἥξει δ᾿ Ὀδυσεὺς ὅσον οὐκ ἤδη

πῶλον ἀφέλξων σῶν ἀπὸ μαστῶν

ἔκ τε γεραιᾶς χερὸς ὁρμήσων.

ἀλλ᾿ ἴθι ναούς, ἴθι πρὸς βωμούς,

[ἵζ᾿ Ἀγαμέμνονος ἱκέτις γονάτων,]    145

κήρυσσε θεοὺς τούς τ᾿ οὐρανίδας

τούς θ᾿ ὑπὸ γαίας. ἢ γάρ σε λιταὶ

διακωλύσουσ᾿ ὀρφανὸν εἶναι

παιδὸς μελέας ἢ δεῖ σ᾿ ἐπιδεῖν

τύμβωι προπετῆ φοινισσομένην    150

αἵματι παρθένον ἐκ χρυσοφόρου

δειρῆς νασμῶι μελαναυγεῖ.

145 del. Heimsoeth 147 γαίας Porson: γαῖαν mss. 150 τύμβωι recc.: τύμβου veteres⁹²

Odysseus will come almost at once to pull the filly away from your breast and hurry her from your aged embrace. Come, go to the temples, go to the altars, [sit as suppliant at Agamemnon's knees,] loudly invoke both the gods of heaven and those beneath the earth: for either prayers shall prevent your being orphaned of your child or you must live to see the maiden thrown forward on the tomb, crimsoned with the blood welling dark and gleaming from her gold-decked throat. (tr. Collard (1991) adapted)

Both Janko and Bremmer noted the verbal similarities but did not explore the importance of the context. There are several points of contact. Polyxena is compared to an animal lead away from the breast of the mother, a rather incongruous image if one thinks of the age of Hecuba. This is very similar to the image of the goat who has copious milk and is lead away from the garden; it also recalls the ‘goat rushing to the milk’ in the so-called Orphic Golden tablets.⁹³ Both the Hecuba passage and the Getty Hexameters envisage the necessity of prayers to avert evil and death. The adjective μελαναυγεῖ ‘“dark-shining” is a vivid description of the gleaming of a dark liquid, imitating the epic phrases μέλαν αἷμα (536–7n.), αἷμα κελαινόν (Il. 1.303), αἷμα κελαινεφές (Il. 4.140)’.⁹⁴ As Janko notes:

The non-traditional word μελαναυγής is first attested in Euripides’ Hecuba 152, νασμῶι μελαναυγεῖ. It is very striking indeed that νασμός occurs here in the very next line, as if its composer knew this passage of Euripides (the word does indeed seem like one of the latter's creations) – unless of course Euripides is creating a tragic variation on the phrasing of this very poem, a stream of blood instead of a stream of milk; for μελαναυγής seems more obviously applicable to a place in the underworld than to blood.⁹⁵

This works much better if it is Euripides who is offering a variation on the passage present in the Getty Hexameters. Blood alludes to death and the afterlife, and the ‘stream’ of blood recalls the filly/breast image of Hec. 142, and the milk of the goat in GH 11.

In fact, Euripides uses similar oxymoronic images already in the Alcestis, in another passage which recalls the language of the Getty Hexameters (Alc. 259–62):

ἄγει μ' ἄγει τις, ἄγει μέ τις (οὐχ

ὁρᾶις;) νεκύων ἐς αὐλάν,

ὑπ' ὀφρύσι κυαναυγέσι

βλέπων πτερωτὸς Ἅιδας.

Someone is taking, is taking me (don't you see him?) away to the court of the dead. It is winged Hades, glowering from beneath his dark brows. (tr. Kovacs (1994))

The adjective κυαναυγέσι (‘dark gleaming’), which describes the eyebrows of Hades, is a variation on μελαναυγέϊ (‘dark gleaming’) used in Hec. 152 in reference to the afterlife.⁹⁶ The form κυαναυγέσι transfers to the eyebrows the gleaming of Hades’ eyes. The repetition of ἄγει and the mention of the ‘hall of the dead’ recall the ‘garden of Persephone’ and the verb ἄγει of GH 9. The oxymoron for the description of Hades is frequent in tragic language: see Soph. Aj. 394a–5 Ἰὼ | σκότος, ἐμὸν φάος, | Ἔρεβος ὦ φαεννότατον ‘Io darkness, my light, o Εrebus, most bright for me’⁹⁷ and Eur. Hel. 518–19 μελαμφαὲς … ἔρεβος.⁹⁸ The ‘light’ of Erebos is in fact darkness.⁹⁹ As mentioned at the beginning of the article (above, section 1), the fourth-century BC tragic poet Carcinus alludes to this traditional language in his description of Persephone's journey to the Netherworld when he talks of γαίας … μελαμφαεῖς μυχούς ‘earth's depths, whose light is darkness’ (70 F 5.3 TrGF; tr. Oldfather (1939)).

The occurrence of so many similar phrases in reference to the oxymoronic ‘dark light’ of Hades or Erebos makes it likely that the adjective μελαναυγής, like μελαμφαής and κυαναυγής, was created in reference to Hades/Erebos, possibly in cultic poetry, and that the phrase in GH 8 μελαναυγέϊ χώρωι uses it in the original context.

As Scott Scullion notes (personal communication), Euripides uses these adjectives in a pointed oxymoronic way, whereas the Getty Hexameters seem to use μελαναυγέϊ in the straightforward meaning ‘black’, in contrast with the light of the ‘bright goddesses’ (GH 12 θεαῖς πεπιθοῦσα φαειναῖς) and their ‘torches’ (13 ⸤λ⸥αμπάδας). This would point to Euripidean effective manipulation of traditional magic and cultic language.

Polyxena, in the same scene, is compared to a ‘whelp’ (Hec. 205 σκύμνον) and to a ‘heifer reared in the mountains’ (205–6 οὐριθρέπταν | μόσχον), which will be killed and go to Hades.¹⁰⁰ This again recalls the goat coming ‘down from the mountains’ for milking mentioned in GH 8. Later in the play, the sacrifice of Polyxena is described in terms that clearly allude to the narration of the sacrifice of Iphigenia in the Agamemnon.¹⁰¹ No ritual perspective is open for the chorus of Hecuba in the time frame of the play, which is set in a place where no divine cult is performed. For chronological and poetic reasons, it is unlikely that the Getty Hexameters reworked phrases from Euripides (and they definitely do not allude to Euripides). In Euripides the ‘linguistic intimation’ of texts on the afterlife is stronger in the Alcestis than in the Hecuba. On the other hand, Carcinus is clearly alluding to the language of mystery cults (‘linguistic intimation’) and possibly to the very phrase used in the Getty Hexameters (and other similar early texts?) (‘textual allusion’).

9. The Chaldaean Oracles, Synesius and Iamblichus

The phrase had a long and successful life in late antiquity. The adjective ‘dark-lit’, Bremmer notes, ‘occurs otherwise only in the Chaldean Oracles (fr. 163 des Places: with thanks to Radcliffe Edmonds) and the Orphic Argonautica (513)’.¹⁰² Bremmer does not explore the relation between these texts and the Getty Hexameters, but it is easy to observe that these occurrences prove that the text of the Getty Hexameters was read and imitated in the Imperial age. We know for certain from an Egyptian papyrus, K, that the lines about the ‘dark-lit place’ were still copied in the second or even third century AD.¹⁰³ Several late antique passages appear to be clear textual allusions to the phrase found in the Getty Hexameters (and in other texts). The phrase about the ‘dark-lit place’ from the magic tradition is an exemplar-model for the Chaldaean Oracles, and that in turn (and in combination with the phrase from the magic tradition) is an exemplar-model for passages in Synesius and Iamblichus.

Fragment 163.1–3 from the Chaldaean Oracles runs as follows:

Μηδὲ κάτω νεύσηις εἰς τὸν μελαναυγέα κόσμον

ὧι βυθὸς αἰὲν ἄμορφος ὑπέστρωται καὶ ἀειδής,

ἀμφικνεφὴς ῥυπόων εἰδωλοχαρὴς ἀνόητος

Do not stoop below into the dark-gleaming world beneath which an abyss is spread, forever formless and invisible, dark all around, foul, delighting in images, without reason.¹⁰⁴

We should keep in mind the fact that the speaker of at least some Chaldaean Oracles is the goddess Hecate.¹⁰⁵ The Chaldaean Oracles are normally dated to the second century AD;¹⁰⁶ this fragment is transmitted by Synesius, On Dreams 7.5, as well as by Psellus, Damascius and Proclus.¹⁰⁷ The ‘dark-gleaming world’ is the sublunar world, from which the souls should be able to detach themselves, and the ‘abyss’ is Tartarus.¹⁰⁸ The adjective implicitly compares the sensible world with the afterlife. The Platonic image of the cave also contributes to the idea that the world is a ‘dark-lit place’, and that light only shines in the ideal world of the Forms.

Synesius comments on fr. 163 in the context of a discussion of the afterlife. He claims that ‘the oracles’ (i.e. the Chaldaean Oracles) tell that dreams are comparable to the images seen in the afterlife (7.2); he quotes Plato's Laws 653a and Heraclitus (7.3), and discusses the separation of soul and body at death and the importance of mystery cults for the soul (7.4). He then quotes the ‘sacred logoi’ (i.e. the Chaldaean Oracles again) and discusses Lethe and the descent of the soul towards the material world. He quotes again the Chaldaean ‘logos’ in chapter 9, this time in a form that clearly recalls that found in the magical tradition of the Getty Hexameters (Synesius, De insomniis 9.1):

ῥεψάσης μὲν κάτω ψυχῆς, ἔλεγεν ὁ λόγος ὅτι ἐβαρύνθη τε καὶ ἔδυ, μέχρις ἐγκύρσηι τῷ μελαναυγεῖ καὶ ἀμφικνεφεῖ χώρωι

when the soul descends, the logos says that it becomes heavy and plunges down, until it reaches the dark-shining and all-around-dark place.

Synesius seems to have known the phrase as attested in the Getty Hexameters. He imitates the Chaldaean Oracles (not the Getty Hexameters) in his first Hymn, lines 297–301:

ψυχά τ' ἀκλινὴς

καὶ κλινομένα

ἐς μελαναυγεῖς

χθονίους ὄγκους

Both the soul that is not inclined (towards the material world) and the soul that is inclined towards the dark-gleaming earthly material substance (sing the praise of the Lord).

Here Synesius glosses the phrase with the adjective χθονίους ‘earthly’, so as to make sure that the adjective is understood (as the riddling text of the Chaldaean Oracles implies) to refer to the world we live in, not the afterlife (as in the magical tradition).

One last Neoplatonic/Neopythagorean imitation of the phrase is found in Iamblichus, On the Pythagorean Life. Iamblichus claims that Pythagoras presented himself as a divine man and that other human beings, being unable to look directly at the gods, should look at him to understand the nature of the divine. Similarly (67),

when people cannot look directly at the sun, because of the brilliance of its rays, we find ways to show them an eclipse, with a deep container of water or a film of pitch or a black-backed mirror (ἢ καὶ διὰ τετηκυίας πίσσης ἢ κατόπτρου τινὸς μελαναυγοῦς), sparing their weak eyesight. (tr. Clark (1989))

Pythagoras himself is the ‘black-shining’ mirror that allows humans to see the gods.¹⁰⁹ This passage clearly alludes to Plato's cave myth (humans cannot look at reality), to Plato's comparing imitation with mirrors (Resp. 596d–e) and to the myth of Er (Pythagoras himself descended from heaven).¹¹⁰ Pythagoras thus can provide humans with access to the divine, and help them avoid descending into base types of life.

These passages show how the phrase describing the ‘black-shining places’ survived for a long time in poetry and prose that discussed the fate of the soul and mystery cults. It is likely that both the author of Chaldaean Oracles fr. 163 Des Places and Synesius knew the version of the text attested in the Getty Hexameters and in papyrus K.

10. Conclusions

The complex allusions of tragedy to magic texts (‘linguistic intimations’) suggest that some linguistic elements that surface in the Getty Hexameters (esp. GH 8 μελαναυγέϊ χώρωι, 26–7 ͱότα[ν κὴρ] | [ἄφνω ἐπ’ ἀ]νθρώποις θανατηφόρος ἐγγύ[θεν ἔλθηι], 31 [βέλτιον ἔ]σ{σ}τι πόλει) where known and imitated in Greek tragedy and by Plato.

The Getty Hexameters, though, are not (as far as we know) an authoritative, stable text, nor were they attributed to a real (Pythagoras)¹¹¹ or fictional (Orpheus? Musaeus?) author. Sections of the texts that surface in the Getty Hexameters were being used and adapted in various forms in the classical age. Some tragic and philosophical texts allude to the (until recently unknown) sub-genre to which the Getty Hexameters belong (incantation-paean in hexameters) (‘genre intimation’) and to specific phrases (‘linguistic intimation’ and, possibly, in one case, ‘textual allusion’) used in this sub-genre. The tragic performance presents itself as reperforming (or attempting to reperform) on stage the incantation rituals that it alludes to. It is in the nature of the tragic text to englobe discourses and speech-genres belonging to different traditions; it is also in its nature to stress the ritual failures of chorus and characters.¹¹² It is unlikely that the ‘short story’ about the goat of GH 8–14 alludes to/depends on the text of the Hecuba (no ‘textual allusion’ here). The similarities in language probably derive from a phrase used of Hades/Erebos in the Getty Hexameters, a phrase which was known to Euripides, and that was going to be in continuous use from (at least) the fifth century BC to the fourth century AD, resurfacing in authors who discussed the afterlife and mystery cults. This new antecedent of tragic texts shows how dense the web of allusions (‘genre intimations’, ‘linguistic intimations’, ‘textual allusions’) in the works of the tragic authors is, and how much richer the interpretation becomes whenever we acquire new material that expands our knowledge of genres, ritual practices and language in classical antiquity.