1.7: Stefan Hagel Teaches Us How to Sing

Guest Feature 1

© 2021 Philip S. Peek, CC BY 4.0 https://doi.org/10.11647/OBP.0264.07

Singing Ancient Greek by Stefan Hagel

Why would you want to do such a weird thing? Perhaps because you realise most of ancient poetry was meant to be sung, and you would like to experience it as properly as possible. Or because it is a nice way of getting acquainted with the interplay between rhythm and melodic accent that is at the core of ancient Greek pronunciation. Or because, in the form of song, it is much easier to remember poetry. Or to impress your friends (only recommended with a very special type of friend).

If you do not like singing and are nevertheless fascinated with poetry, there is still much to gain—epic poetry was generally recited at least from the Classical period on, and so were all the ordinary dialogues in drama. Here the rhythm alone, if executed in the ancient way in the form of long and short syllables, will grant a genuinely musical experience even without a melody, and all the more so if the gliding pitch contours of the language are respected as well. For those, however, who take the easier route of singing, here are some ideas.

Regarding the music of ancient poetry, we are trapped between the excitingly ancient and the reassuringly genuine. We do have some ancient tunes, some of them virtually complete, but most of these are from the Roman period, half a millennium after Classical Athens and even further removed from the music of Sappho and Homer. Most famous among these melodies is the so-called Seikilos song, inscribed on a small funeral monument. Then there is a number of pieces by a certain Mesomedes, a highly valued musician at the Imperial court in Rome. Apparently composed for elementary music schooling, they are still perfect introductory material, having been preserved in medieval copies. The oldest performable melody, in contrast, comes from the end of the second century BCE, when a hymn staged at Delphi by the Athenian state orchestra was publicly displayed on stone. Naturally, this is much more complex music, which will sound quite strange to ears not trained in the subtlety of ancient chromaticism.

The aforementioned melodies are readily accessible in modern transcriptions. There is only a small problem: most of these do not reflect the original pitch. When the difference is in the range of only a semitone or tone, that may be negligible for many purposes, but often—which unfortunately includes the best editions—the melodies appear much too high. This has complex historical reasons that need not concern us, but it is important to bear in mind. For the simpler pieces mentioned above there is a simple guide: the Seikilos song, Mesomedes’ Hymn to the Sun, and the two short Invocations in the same collection all start with the same note. However we find it transcribed (most often as a or e), its original pitch was very close to modern F#. Remember this note, in case you should ever want to play one of the chief ancient instruments, where it is a traditional bass note (in ancient musical notation, it is written as C).

Even though the Greeks developed the idea of an octave being divided into twelve equal semitone steps, they did not normally tune their instruments in this way. Therefore, an original performance would often surprise modern Western ears with some notes slightly ‘out of tune’. If you are interested in such details, I have developed tools for bringing transmitted tunings to life on the computer and experimenting with them. You may download them here,

https://homepage.univie.ac.at/stefan.hagel/software/Harmogai_inst.exe

Finally, you may want to set your favourite lyric to melodies of your own. The good news is that here we cannot do anything wrong, except a single thing: to claim that what we are doing would amount to ‘reconstructing ancient music’. Otherwise we enjoy complete freedom, though it is always helpful to remain aware of what exactly we are aiming at. In an approach towards the unapproachable goal of ‘authenticity’, there are various possible steps—challenges best faced one at a time: trying to understand the underlying rhythm, rationally and as a bodily experience; roughly using ancient-style scales; perhaps using real ancient tuning; using reconstructed instruments (in recent years, good replicas of auloi and lyras have become available, though the best-advertised are not necessarily the most authentic); and of course designing the melody in an ancient way.

Unfortunately, our understanding of what the last may mean rests on the few surviving music fragments plus some scattered remarks in ancient texts. Above, we have already made the acquaintance of a typical starting note, which also serves as a convenient final; we also learn that the note a fourth above it played a highly important role in pre-Roman music, and the notes a tone below these two respectively may conveniently serve as a harmonic contrast. An instrumental accompaniment would underline such harmonic domains, while abstaining from modern chordal harmony: instead of our cherished chords involving three or more different notes in the octave, the ancient sources only ever talk about a combination of two, creating intervallic concords but also discords, as well as resolving the tensions created in this way into unison.

But how to develop the contours of a melody? In many of the extant ancient pieces, there is a tendency to follow the rise and fall of speech melody, sometimes roughly, sometimes so faithfully that scholars were able to derive crucial parameters of ancient prosody from them. Many of us who have composed melodies for ancient texts have therefore mimicked this practice, if only for lack of a better starting point. Probably this approach is indeed well suited for many kinds of poetry; it has been argued that even early epic song worked in a similar way. But this will not help us with most strophic song. Whenever the ancient Greeks composed in strophes, they replicated the rhythm in each of them with great precision, while taking (almost) no heed of where the accents were placed in the verses: in sharp contrast to English, for instance, Greek accents were purely melodic and thus did not contribute to the rhythm of the language. As a consequence, if ancient Greek poetry performed matching strophes to similar melodies, these could not possibly all follow the contours of speech melody. One may doubt the condition—might they rather have varied the melody from strophe to strophe? Unfortunately, the extant musical fragments contradict that possibility at least for the songs of tragedy. Still, the matter may have been different with the much shorter strophes of Sappho, Alcaeus, and other archaic composers. Here the frequent repetition of a small structure may well have called for some melodic variation—especially when a single performer could easily adjust melodic patterns on the fly, which a chorus cannot.

The guidelines for ‘composing according to the accent’ are rather straightforward. In each Greek word, apart from small ones such as prepositions and articles, there is a certain point where the gliding voice of speech reached the highest pitch. This point is indicated graphically by the accent marks: an acute or grave was located at the end of the vowel or diphthong, with a circumflex, rather than at the beginning. After this point, the pitch dropped markedly—except in the case of a grave accent, where the end of the word followed suit, leaving no time for a pitch drop. From the start of the next unit up until its own accent point, the pitch does not fall again. Notably, there is never the need for a sharp rise—it is the downward movement that defines the accent preceding it.

Converted from the continuous glide of the speaking voice to the individual notes of song, this means: (1) within an accentual unit, the pitch should not fall before the accent is reached. (2) If possible, fall immediately after it. In the extant melodies, syllables with circumflex often bear a couple of descending notes (or even more), but often also a single note or even a rising movement, followed by a drop on the next syllable. On the sentence level, melodies tend to fall gradually, being reset to a higher level after a phrase boundary; this is just a typical feature of human speech. Words bearing emphasis, including proper names, may also be elevated to higher pitch, relative to their context. It is worthwhile observing how the extant melodies bear these ‘rules’ out (or don’t); for us, they may provide a conveniently tight framework within which musical creativity can unfold.

All this gives us a toolkit for raising ancient poetry from the dormancy of printed form to the auditory life for which it was meant. The journey, however, does not end here. There was chromaticism, opening up potentials of modulation unheard of in Western music. Most excitingly, perhaps, the music of the Classical period involved the so-called harmonía, incorporating microintervals down to quartertones, typically played on kinds of doublepipes which music archaeology has only just begun to make sense of. Lots of fun still lies ahead.

Practice Learning How To Sing

Read the first line of the Iliad, preferably memorizing it.

μῆνιν ἄειδε θεὰ Πηληϊάδεω Ἀχιλῆος

Next listen to the singing of this line, paying attention to the high notes and the lower notes,

Stefan Hagel Sings the Iliad.1

Each accented vowel has a high pitch after which there is a falling off or a lower pitch. For μῆνιν the high point and fall occur on the eta. For ἄειδε the high point is on the alpha and the fall is on the diphthong ει. For θεὰ there is a rise to the alpha. For Πηληϊάδεω the fall occurs after the alpha. Ἀχιλῆος rises towards the eta and falls within it.

Try the same exercises with the first seven lines of the Iliad, again preferably memorizing them.

μῆνιν ἄειδε θεὰ Πηληϊάδεω Ἀχιλῆος

οὐλομένην, ἣ μυρίʼ Ἀχαιοῖς ἄλγεʼ ἔθηκε,

πολλὰς δʼ ἰφθίμους ψυχὰς Ἄϊδι προΐαψεν

ἡρώων, αὐτοὺς δὲ ἑλώρια τεῦχε κύνεσσιν

οἰωνοῖσί τε πᾶσι, Διὸς δʼ ἐτελείετο βουλή,

ἐξ οὗ δὴ τὰ πρῶτα διαστήτην ἐρίσαντε

Ἀτρεΐδης τε ἄναξ ἀνδρῶν καὶ δῖος Ἀχιλλεύς.

Again listen to Stefan Hagel’s singing of these lines, paying attention to the high notes and the lower notes,

Stefan Hagel Sings the Iliad.2

For comparison, listen to the same text in reconstructed classical Attic pronunciation,

The Beginning of the Iliad Spoken.3

As you practice pitch accent, use this simple strategy to hear the rhythm and melody of the words.

---

1 https://soundcloud.com/stefan-hagel-448623467/il-1-1-100-3.

2 https://soundcloud.com/stefan-hagel-448623467/il-1-1-100-3.

3 https://soundcloud.com/stefan-hagel-448623467/il-1-1-100-spoken.