ancient music, Seikilos Epitaph

We’ve come to the ‘Louie Louie’ of ancient music: a song that exists in uncountable versions.

W. M. Ramsay, a Scottish archaeologist, discovered a marble stele during an excavation at Aydin, western Turkey, in 1883. The stele has inscribed on it a complete song, with words and musical notation, in ancient Greek. It has been dated to around the second century AD, making it the oldest complete song on record. (The music I’ve covered hitherto has been fragmentary, or been widely open to different interpretations.) The stele is now in Copenhagen, in the National Museum of Denmark.


A stele (or stela) is an inscribed upright stone serving as a monument, usually a tombstone. This particular stele is dedicated by a man named Seikilos, and the words of the song are an exhortation to enjoy life and not to grieve, for time will soon demand that all things end.

You could spend hours online listening to the various interpretations of this piece of music, but, unlike the Hurrian Hymns, or the chorus from Orestes, the music is instantly recognisable as the Seikilos Epitaph in every version: the notation is not disputed. There are, however, disagreements about pronunciation. In my last post, I threatened to completely solve the problem of The English. I’ll deal with this in short order. I’m English, and when I hear English singers singing in ancient Greek, it’s pretty obvious to me that they are English. It’s especially obvious if they are affiliated with Oxford university or something like that, because they sound like they’re doing a Received Pronunciation version of Ancient Greek. It’s also worth remembering that modern Greek has a very different pronunciation to ancient Greek. So, to modern Greek ears, I can imagine that the English accent combined with the academic insistence on trying to cleave close to ancient pronunciation, sounds horribly cacophonous. Now, when Greek listeners, whose ears are offended, get behind their keyboards to complain online, the English interpreters, justifiably – with the weight of academic research behind them – get narked. And the whole thing gets ugly, and no-one comes out of it well. Basically, posh English people do sound silly when they put on foreign accents, but THAT’S FINE. Everyone sounds silly when they put on foreign accents. For those offended by any accents, dialects or pronunciation in these versions, may I recommend a clarifying dose of ‘Heartbreak Hotel’ by the Swedish Elvis Presley, Eilert Pilarm

Good. That’s settled. Back to Seikilos.

The earliest modern reappearance of this piece of music of which I’m aware is surprising. The 1951 film Quo Vadis, with Peter Ustinov as Emperor Nero, includes a scene with Nero extemporising a song which happens to have a melody almost identical to that of the Seikilos epitaph (though the words are not the same). Thanks to my pal Tony for spotting this.

From the 1950s, recorded versions begin to proliferate, albeit slowly at first. Now the Internet gives us access to scores, possibly hundreds of recordings. I’m going to highlight three versions.

The Gardzienice Centre of Theatre Practices was founded in Poland in 1977, and, among its varied projects, hosted the Metamorfozy performance piece in the 1990s, and the Ancient Orchestra project from 2001 to 2004. The intention behind these projects was to recover the Dionysian elements in a music which had — they felt — been too politely Apollonian in its reception. (It’s worth exploring their Bacchic interpretation of ancient music at length.)

For the Metamorfozy album (2000), the Seikilos Epitaph was retitled after its initial words, rendered in Greek and Polish:  ‘Hoson Zes / Dopóki żyjesz’

In 80 short seconds, it creates an atmosphere of remoteness and disquietude through a drone, some off-mic vocalisations, and a dissonant flute. The soft, breathy vocal enters the sound-picture like some sexy ghost, disturbing the air just enough to hit the right notes, before the whole thing fades out again.

The music lives, and Athenian ensemble Lyre ‘n’ Rhapsody have taken the material of the Seikilos Epitaph as a starting point for a longer variation (titled ‘Oson Zeis’, also on their album Awakening the Muse, 2013).

It’s pleasing to hear this sung by Athenians with modern Greek pronunciation, merging the ancient with a modern pop and folk tradition. Consequently, the rendition possibly involves over-emoting to an extent that may jar with my revoltingly hip Londoner tastes, but I can’t blame them for exploiting the full melodic potential of their interpretation.

Finally, it’s those English musicians with the weight of Oxford academia behind them that I mentioned earlier. This version is distinguished by the musicianship of master aulos player Barnaby Brown, and practised interpreter of ancient songs, Stef Conner, singing and playing lyre. This performance is extracted from the Radio 3 Early Music Show (listen here at 50’38”).

This interpretation doesn’t waste any time. The rhythm of Brown’s double aulos gives the song such lift at the start, before unobtrusively fading to an accompanying drone for Conner’s voice. When the aulos it repeats the melody line after the singing, it provokes a strong rush of joy before promptly ending, well under a minute. Too short, like the life described by the inscription on the stele.

ancient music, Greek chorus

As we move on in time, the archaeological/historical record gets noisier. It becomes increasingly crowded with evidence, however hard to decipher it may be. I haven’t even mentioned Egyptian, Chinese, or Indian music yet.

Greek tragedies are not plays in the modern sense: they are gesamtkunstwerken: total artworks of stylised speech, chanting, singing, dancing to be performed in a place of geographical, social and religious significance during the City Dionysia festival. To read the text of a tragedy is akin to reading the libretto of an opera: it omits a great amount of vital information.

Consequently, it is always fascinating to hear how artists, scholars, and musicians attempt to recreate the sound-world of tragic theatre. I was alerted to this work in 2006 by a newspaper article by Robert Thicknesse, accompanying a radio programme on the power of a chorus. Thicknesse recruited composer Roxana Panufnik to set the final chorus of Sophocles’s Oedipus Tyrannos (in Oliver Taplin’s translation) to music based on the research of Martin L. West. I missed the resulting programme (In Chorus, Radio 4, 11 July 2006), which now dwells in the “pre-I-Player” archive. This commission, however, was a new piece of work based on ancient techniques, but there are examples of attempts at interpreting the surviving notation from the tragedy of Orestes by Euripides.

The play dates from 408 BC, and a few lines from the first choral ode survive with accompanying musical notes, providing melody, but not rhythm. The earliest interpretation of which I am aware is from 1979, by Atrium Musicae de Madrid (0’19” – 3’02” on this video):

This has all the clanging and dissonance one associates with twentieth-century representations of ancient ritual. The singing is a solemn chant. To my layman’s ear, it’s an evocative meeting of plainsong and minimalism. There are numerous other recordings available, all largely agreeing with this template, although some are closer to modern Greek pronunciation and folk melodies.

Modern Greek pronunciation and folk melodies became a bone of contention for some commentators on the recent scholarly reappraisal of the ode from Oxford University. The critics believe that the British interpretation is compromised for ignoring living folk traditions which, it is claimed, have continuity with ancient traditions. I’m in no position to judge on that, but I will return to problem of Englishness when I come to the Seikilos Epitaph in a week or so.

The scholarship behind this new version is, by all accounts, rigorously investigated and comprehensively evidenced. The major deviation from previous versions is in the rhythm: it is fast and lively, not solemn and stately. The following video is the conclusion of a terrific, brief documentary on ancient Greek music by Armand D’Angour:

D’Angour’s interpretations are discussed further on Radio 3’s The Early Music Show, from 28 January 2018: The Music of Ancient Greece (go to 26’30” for the Orestes ode).

It sounds true to me: remembering that the chorus is not to be heard in isolation, but has been extracted from a longer dramatic work from the festival of Dionysus. It reminds me of the Choephoroi (Libation Bearers) of Aeschylus. I’ve written before that this play, with its long grave-side laments, can be very flat on the page and in performance. But the performance at the Globe in 2015 was the highlight of that version of the Oresteian trilogy. The musical and choreographic direction of the chorus rescued scenes that in other’s hands I had yawned through.

Here, in the Orestes of Euripides, the chorus sings of the ‘swift-running winged ones, goddesses of madness’ — the Furies of Clytemnestra — who torment her murderer: her son Orestes. It is easy to imagine this being sung at the Theatre of Dionysus on the Athenian Acropolis, contributing to the audience’s appreciation of Orestes’ tormented mental condition.



Greek Text (Madrid 1979 version, lines 339, 338, 340-44)

κατολοφύρομαι κατολοφύρομαι.
ματέρος αἷμα σᾶς, ὅ σ᾽ ἀναβακχεύει;
ὁ μέγας ὄλβος οὐ μόνιμος ἐν βροτοῖς:
ἀνὰ δὲ λαῖφος ὥς τις ἀκάτου θοᾶς
τινάξας δαίμων κατέκλυσεν δεινῶν
πόνων ὡς πόντου λάβροις ὀλεθρίοι-
σιν ἐν κύμασιν.

Greek Text (Oxford 2017 version, lines 317-28, 333-44)

δρομάδες ὦ πτεροφόροι
ποτνιάδες θεαί,
ἀβάκχευτον αἳ θίασον ἐλάχετ᾽ ἐν
δάκρυσι καὶ γόοις,
μελάγχρωτες εὐμενίδες, αἵτε τὸν
ταναὸν αἰθέρ᾽ ἀμπάλλεσθ᾽, αἵματος
τινύμεναι δίκαν, τινύμεναι φόνον,
καθικετεύομαι καθικετεύομαι,
τὸν Ἀγαμέμνονος γόνον ἐάσατ᾽ἐκ-
λαθέσθαι λύσσας μανιάδος φοιτα-
λέου. φεῦ μόχθων, οἵων, ὦ τάλας,
ὀρεχθεὶς ἔρρεις,

τίς ἔλεος, τίς ὅδ᾽ ἀγὼν
φόνιος ἔρχεται,
θοάζων σε τὸν μέλεον, ᾧ δάκρυα
δάκρυσι συμβάλλει
χορεύων τις ἐς δόμον ἀλαστόρων
ματέρος αἷμα σᾶς, ὅ σ᾽ ἀναβακχεύει;
κατολοφύρομαι κατολοφύρομαι.
ὁ μέγας ὄλβος οὐ μόνιμος ἐν βροτοῖς:
ἀνὰ δὲ λαῖφος ὥς τις ἀκάτου θοᾶς
τινάξας δαίμων κατέκλυσεν δεινῶν
πόνων ὡς πόντου λάβροις ὀλεθρίοι-
σιν ἐν κύμασιν.

ancient music: Hurrian Hymns

We played music, we wrote down stories, it’s inevitable that we would start trying to preserve music in a written form too. Once the novice starts digging, there is a surprising amount of ancient evidence to be found, even if scholars can’t agree on the interpretation.

The Hurrian Hymn to Nikkal is the oldest surviving near-complete notated music we have. The Hurrians lived in Anatolia and northern Mesopotamia (i.e. modern Turkey, Syria, and Iraq) during the bronze age (in evidence very roughly between middle of third millennium, and the thirteenth century BC). The hymn, along with other song fragments carved on clay tablets in cuneiform script, was discovered in the 1950s during excavations of the ancient city of Ugarit, in modern-day Syria.

Hurrian 6 tablet. (Image credit)

The Hurrian language is not completely understood, and the Akkadian notation is open to interpretation. Consequently, there are many different approaches taken to this piece of music. It is a hymn to the goddess Nikkal who is associated with the orchard, and is wife of the Moon god Yarikh.

Numerous responses to the song can be heard online. Martin L. West, a prolific classical scholar, published a version in 1993, which has been performed by Ensemble De Organographia:

This version, like several others, is marked by a staccato stateliness which effectively evokes the ritual clichés of the twentieth-century pop-cultural attitude to ancient civilisations. I don’t mean that as a criticism by any means — only that it perfectly recreates my imagined (and historically-uninformed) atmosphere of bronze-age court music. But it is beautifully performed, and I recommend the album it appears on. You can find more interpretations along the same lines (one, two, three, four).

It’s no surprise to find that the Hymn to Nikkal has inspired modern variations: among them, a project in New York and a modern orchestral setting by Syrian composer and pianist Malek Jandali. Jandali’s piece owes its melodic inspiration to the recent interpretation by archaeomusicologist Richard Dumbrill — an interpretation which has also been set to an ancient lyre by Michael Levy. In addition to his scholarly publications, Dumbrill has argued for his interpretation across a number of You Tube videos (one of the most recent of which is here). Here, he concludes that the hymn is a song without instrumental accompaniment: ‘an instrumental accompaniment to this song — prayer, as far we understand it — would be incongruous with the private nature of this young woman’s prayer: asking the goddess Nikkal to make her fertile’. He makes a convincing case, and the resulting recording is freed from the staccato style of the other versions. In comparison, these other versions sound tentative and restrained — almost mechanical — as if afraid to allow emotions and feelings to obscure the text. Instead, we hear the song as something far more arresting and human.




ancient music, new lyres

We shared our stories; retold them generation after generation. They lived in the cultural memory. It was the accountants who were writing things down. The book-keepers of Mesopotamia used cuneiform script to keep records (from about 3200 BC). And many such records survive on clay tablets, but, eventually, they were joined by writings which told stories: heroic tales, exploits of the gods, cultural histories — myths. It is the earliest surviving literature we can read today, reconstructed and translated from excavated clay.

Evidence of musical culture has also been recovered: instruments and images of musicians. But we have no music.


12 clay tablets containing the Epic of Gilgamesh, from Nineveh, 7th century BC.
Now in the British Museum. Photo: Ben Pestell

The recent Ashurbanipal exhibition at the British Museum, as you might expect from an exhibition on the life of the king of the world, no less, focused on the great man, war, grand spectacle, and so on. Aspects of social history, however, were evident in the many examples of tablets from the library of Ashurbanipal. Moreover, the lions and hybrid human-animal figures depicted on reliefs and giant sculptures gave a sense of a life lived closer to nature: where the ontological boundaries between all things are permeable. This is the world of myth. King Ashurbanipal, reigning from 669 to around 631 BC, governed a society still negotiating one of the problems of Gilgamesh and Enkidu: the transactions between our human and animal natures which result in profound loss of each side.

Court musicians playing horizontal harps. Nineveh, seventh century BC.
Now in the British Museum. Photo: Ben Pestell

I was particularly interested in the occasional images of musicians. Though we don’t have a reliable means of reconstructing the music of ancient Mesopotamia, we do at least have the means of reconstructing some instruments. The most famous instruments are the Lyres of Ur: the Gold Lyre, with its gold bull’s head, and the Silver Lyre, completely covered in silver.

There are at least three replicas of the Silver Lyre in use today. One is by Richard Dumbrill, co-founder of the International Council of Near Eastern Archaeomusicology (ICONEA), and one of the curators of the Ashurbanipal exhibition. His improvisation on the lyre (which he claims would not have sounded incongruous to third-millennium-BC Sumerians) can he heard at the start of this video. Another is by Canadian musician Peter Pringle who has information, and his own improvisation, on his website.

An especially ambitious project is by The Lyre Ensemble (using lyres built by Jonathan Letcher). The Lyre Ensemble sets ancient texts (in Babylonian and Assyrian, and in English translation) to the music of three replica lyres (Gold, Silver, and the smaller Pharaonic Lyre). The ensemble’s CD, The Flood is an album of evocative new music by modern Europeans, but with reconstructions of ancient instruments and with ancient texts.

At times, The Flood cannot help sounding distractingly English, perhaps a little mannered, and with much of the music minimal and repetitive, the voice carries the lion’s (or the bull’s) share of each song’s emotive weight. The overall impression of the album is of a solemn and pared-back folk music.

The Flood brings out some shades of social history in the settings of brief proverbs, a lullaby, a love song, and a song to a mother, while also singing on mythological themes of the great flood, the civilising of Enkidu, the descent of Ishtar to the underworld.

Among the highlights are ‘Enkidu Curses the Harlot’ and ‘Ištar’s Descent’. The former takes advantage of the natural resonance of the lyre’s body to create a creaking-bed-spring metallic bass line.

The Descent transmits a tense forward momentum from the lyre’s taut strings, which creates a compelling backdrop for Stef Conner’s voicing of the katabasis: Conner’s usually clear and light voice is augmented by an assortment of growls and screams to accompany the eschatological moment — of a goddess at the very limits of existence. Although the lyres (played by Andy Lowings, recorded and assisted by Mark Harmer) may be the stars of this disc, it is Conner’s voice which brings them to life, whether in English or in Sumerian or Babylonian. If her phrasing sometimes sounds very familiarly English (she has sung with The Unthanks, steeped in English song), it is no bad thing to be reminded of the temporal and cultural distance between the tools (texts and lyres) and the players.

The great success of The Flood is in bringing these sounds and stories into the digital present, even as — by necessity — it avows its distance from the world it evokes. As such, it feels like the essential and compelling first step in expanding our understanding of what can be done with these ancient tools.

ancient music, archaeoacoustics

It began as a cry… ‘a cry expressing an urge and appeasing it at the same time.’ The first song: a performance of contradictions. So Ernst Bloch decided. Bloch looked far back into the past to discover the first instance of ‘musical differences’. He distinguishes between the ritual drum which accompanied these primal songs, and the pan pipe. The function of the pipe, Bloch argues, was not ‘to induce a stupor or to work magic like the wooden clapper, the cymbal or the magically painted and in itself magically venerated drum. Rather it was confined — pure entertainment apart — to amorous longing and the enchantment of love, the latter being a survival of magic’ (Bloch, Essays on the Philosophy of Music [1974], p. 196). For Bloch, the invention of the pan-pipe is of great consequence for its ability to make a ‘well-organised series of notes’, thus marking ‘the invention of music as human expression‘ (197).

Ernst Bloch (rarely photographed sans pipe)

Since Bloch wrote, many more prehistoric flutes have been discovered, some arguably also associated with Neanderthals. The oldest bone flute yet discovered is thought to date back more than 43,000 years (the ‘Divje Babe Flute‘, discovered in modern-day Slovenia).  To put that in context, it’s roughly contemporaneous (give or take a few thousand years) with the earliest known figurative rock paintings. Despite the efforts of archaeoacoustic researchers to recreate the prehistoric sound-world, it’s a long, long wait until we get to a point where we can confidently interpret a complete piece of musical notation (the Hellenistic Seikilos epitaph, dated to c. AD 100), and then just a short hop to the invention of sound recording in the second half of the nineteenth century. (On which see the First Sounds website for recoveries of the earliest sound recordings.)

Divje Babe Flute (thought to be the bone flute played by Neanderthals)

There are some fascinating archaeoacoustic projects out there which try to give a sense of the experience of palaeolithic, or other ancient musics. Here is an article accompanying the 2015 CTM Festival (Festival for Adventurous Music and Art) on recent work in archaeoacoustics. One of the participants of the 2015 festival was Rupert Till, who does a lot of work in ancient music. Dr Till is one of the people behind the Songs of the Caves website, where you can explore the sounds and prehistoric paintings of the caves of northern Spain. You can also hear audio and see a video on the EMAP (European Music Archaeology Project) website.

Who knows what sort of stories or rituals accompanied the paintings of hunted animals, or the music of the flute? Bloch recalls Ovid’s retelling of the myth of Pan and Syrinx the nymph (in Metamorphoses I). Syrinx is transformed into reeds to escape Pan’s desire, but Pan is then enchanted by the sound of the wind through these reeds. Sealing reeds of unequal length together with wax to form his pipe, Pan thus communes with the vanished Syrinx. In this tale, Bloch sees ‘something contradictory and utopian …. The vanished nymph has remained as sound … and sings in the face of want’ (197). This same want for the vanished people in the cave is the same that is expressed, and speculatively calmed in the archaeoacoustic performance. The vanished prehistory returns, only to insist upon its impossible distance from us.

Gerard Hoet, Pan and Syrinx (c. 1700) (Dulwich Picture Gallery)