Armand D’Angour of Jesus College Oxford looks at the possibilities of recreating the music which the Greeks enjoyed.

‘Oxford Scholar cracks the secret of ancient Greek music!’ was the message posted on Twitter announcing an article published in April 2018 in the magazine History Today. What had happened? For over a hundred years, the nature of ancient Greek music seemed to pose an insoluble mystery: what did it really sound like? Descriptions and musical analyses abound, and an ancient table of the notations used to represent musical notes is available; while some dozens of notated documents on stone and papyrus have emerged since the late 16th century to provide clues to melodic reconstruction. But there was an apparently insuperable issue: the earliest papyrus documents notated with Greek melodies, representing music appeared to come from the Golden Age of Athens, suggested that the music was quite alien to our understanding and sensibility. In particular, it featured the presence of notes that were smaller than a semitone, of a kind that investigators in recent times could pitch only with the aid of electronic instruments, with results that sound alien and other-worldly. The ancient Greeks must have had a different aural apparatus to ours to be able to hear and enjoy such tiny intervals.

The Euripides papyrus

A moment’s thought makes it obvious that such a claim is absurd. Cultural differences notwithstanding, the capacities of human ears have not changed. Moreover, the continuity of many aspects of Greek with that of later European culture would seem to tell against a lack of musical connection; and in fact that large sections of notated music dating from the 2nd century BC (such as the Delphic Paeans inscribed on stone) to the 2nd century AD (such as the ‘Song of Seikilos’) could be and have been realised and performed in a way that make undoubted musical sense to modern ears. The musical idiom of Greece is unlikely to have changed dramatically over the centuries prior to the earliest of those pieces, and there is no historical testimony that suggests that it did.

In 1886, however, the German philologist and musical scholar Rudolph Westphal had set his imprimatur on the topic: ‘The Greeks’ non-diatonic music that admits intervals smaller than a semitone, which are wholly foreign to the modern art, will probably, alas, remain for ever an enigma to scholarship’ (Griechische Harmonik und Melopoeie, ix). A discouraging conclusion; and when I first lit upon the scholarship on ancient music as a schoolboy and musical student in the 1970s, my initial excitement soon turned to disappointment. It seemed that we had to make do with a few scrappy fragments indicating no more than a single melodic line, with weird intervals and gaps, and a plethora of scholarship on modes, ethos, and metrical technicalities. Yet more depressing was the fact that music was supposed to have been of extraordinary value and importance to the Greeks; nearly all ancient poetry from 700 BC (the age of Homer), through the lyric period (Sappho, Simonides, Pindar et al.), to the choruses of Greek tragedy and comedy (5th century BC) was sung to melody and accompanied by instruments. Yet the available evidence would surely never lead us to hear anything remotely as interesting as the simplest sonata by Mozart. I reluctantly accepted the statement quoted by a musical investigator in 1932: ‘Nobody has ever made head or tail of Greek music and no one ever will. That way madness lies’.

This defeatist attitude had an unwelcome personal consequence for me. When I sat the All Souls Prize exam in 1983, I was pleased to be shortlisted, and as was the custom I attended an interview with a large group of Fellows. One of these was the philosopher Isaiah Berlin, who was impressed that I combined both philological and musical pursuits, and eagerly asked me if my research project would finally aim to illuminate ancient Greek music for scholars. With all the confidence of youth I indicated that such a study was hopeless, declaring ‘The evidence is too slight for an extended investigation’. My words elicited murmurs of disapproval, with one Fellow coldly remarking ‘All the more reason to do it, I would have thought’. I failed to win a Prize Fellowship.

More importantly, however, my confident statement was soon proved to be wholly mistaken. A few years later the great Oxford scholar Martin West published his clear and comprehensive Ancient Greek Music (OUP 1992). It was an enormous advance for investigations of the subject. West rightly dismissed Westphal’s dictum, noting that ancient Greek music ‘is better understood by putting it in the broad category of ethnic music, extending down to the most primitive and limited manifestations of the melodic instinct, than by looking in it for the workings of supposedly natural and universal principles which are actually abstracted from German music of the eighteenth and nineteenth centuries’ (3). In transcribing the musical documents accurately into modern notation, West did a great service to musical investigators. One could finally sing and play many of the transcribed pieces with relative ease, allowing for a sense of the musical idiom to become familiar with repetition.

Although West’s book did much to elucidate the ‘enigma’ for scholars, a practical stumbling-block remained: what were those intractable microtonal intervals in the earliest musical texts? What sort of music was this, and how could it be realised and heard? Its apparent ‘alienness’ was one of the reasons that historians of music had long dismissed the notion that Western European music had its roots in ancient Greece and Rome. Instead, the consensus was that the European musical tradition derived from Gregorian plainsong of the 9th century AD, which in turn allegedly derived from the music of Hebrew liturgy. Accordingly, more than a thousand years of Greco-Roman music seemed to have left little trace on European music—in retrospect an extraordinarily unlikely notion, but one that has barely been questioned by musicologists.

In 2013-14 I was awarded a British Academy Fellowship to pursue research into the subject. Among other things I travelled in Greece and Italy, observing and listening to folk traditions (such as the Sardinian launeddas, a triple-pipe) that had their roots in ancient musical practices. At this time another project, European Music Archaeology Project, was producing instrumental replicas from archaeological remains of ancient pipes and pictures of ancient lyres. I met some extraordinary researchers and practitioners, including the pipers Luigi Lai,  Barnaby Brown and Callum Armstrong, and the classicist-musician Stefan Hagel, all of whose work made me more familiar with the instrumental resources and sounds being made available to scholarship.

My own principal insight emerged in 2016 when I was trying to explain to a friend the difficulty of hearing or pitching a quarter-tone or third-tone interval. I observed that if one sang a note to the mid-vowel ‘ah’ and simply changed the vowel sound to a high vowel such as ‘ee’, one was already going to increase a pitch by a quarter-tone – a phenomenon known as ‘intrinsic vowel pitch’. Professional singers have to learn not to allow vowel sounds to alter the pitch of the note, via a technique known as ‘vowel pitch modification’; and I had long suspected that the tendency of the choruses of Attic tragedy to use the so-called ‘Doric alpha’ – Athānā rather than Athēnē and so on – was a musical rather than dialectal choice.

At that point it struck me that the earliest notation of ancient Greek texts might have reflected an attempt to pay technical attention to such small pitch changes. I rushed away and made a thorough analysis of the earliest substantial papyrus, that of the chorus from Euripides’ Orestes of 408 BC. Sure enough, all the microtone intervals notated on that papyrus related to a change of pitch between high, mid and low vowels, or could be interpreted as an attempt to mark the creation of a small differentiation between the syllables of any particular word. The obvious corollary was that the basic harmonic structure of the piece was no less tonal than a piece that employed diatonic (whole-tone) or chromatic (semitone) structures—something that the ancient sources had always implied.

The challenge was then to create a working score of the Orestes chorus based on this insight. Scholars had noted that the existing fragment seemed to revolve around two tonal centres a fifth apart. My reconstruction was helped by the assumption that some repeated words in the text were likely to be set to the same or a similar pattern of melody, and by the precious and rare indication of aulos notes on the papyrus (using an instrumental notation that differs from the vocal one) indicating what we would call a harmonisation of the melody. My special area of expertise had long been ancient rhythm; taught in terms of ‘metre’, this tends to employ an offputting technical terminology, but my own teaching emphasises the way we can hear the rhythms of ancient phrases by understanding their metrical basis.

One of the strange errors made by many who had previously sought to realise the music of the Orestes papyrus was to perform it to a slow, ponderous tempo. In fact both the context and the nature of the metre (‘dochmiacs’) made clear that it should be understood as having had a brisk, agitated rhythm.  An ancient commentator had also noted that at the climax of the verse the chorus declaimed rather than sang a phrase (‘terrible toils’)—a striking effect known in modern music as Sprechstimme. By reconstructing melody, instrumentation, rhythm, and tempo, what emerged was a piece of wonderful, impressive, and moving music. It would have been sung as part of the performance of Euripides’ tragedy by a choir consisting of fifteen men, some of whose voices would have higher than others by an octave. For practical reasons I assembled a chorus of men and women, and with my colleague Tosca Lynch (a specialist in ancient Greek music with unrivalled technical expertise) as musical director,  put on a live performance at the Ashmolean Museum in July 2017. The occasion was filmed and a 15-minute video posted online in December 2017. Within weeks it had reached 100,000 views, and the number mounts by hundreds every week. Watch the video for yourself at:

What the video presents are two pieces of reconstructed ancient music, the Euripides’ Orestes chorus (408 BC) and the Delphic Paean of Athenaeus (127 BC), that are part of a musical tradition which is far from alien to our ears. It also shows Stefan Hagel singing a melody, to the accompaniment of the kithara, that speculatively reconstructs the notes of a melody of a kind that may have been sung to accompany the text of Homer as early as 700 BC. Taken together, these three realisations span a period of over five hundred years, and show that the musical idiom of ancient Greece over that time was by no means alien to modern sensibilities. Analysis of the existing scores shows such practices as the use of a falling melody to indicate dejection and an interval leap to accompany the notion of ‘leaping’. This kind of mimetic use of melody is not universal (it is not found, for instance, in Middle Eastern or Far Eastern traditions), but is very much a feature of our European musical tradition.

The stone with the Delphic Paean performed in the video

The upshot of these discoveries is that scholars are finally in a position to claim that ancient Greek music is likely, after all, to be at the root of Western music. Inherited by the Romans, this kind of music will have been sung throughout the cities of the Empire in the early centuries of the Christian era, as shown by pieces that survive attributed to Hadrian’s freedman Mesomedes (2nd century AD). It would also have provided the harmonic and melodic idiom for the earliest Christian hymns sung in churches and congregations: one surviving Christian hymn from around 300 AD, now in the Sackler Library in Oxford, is the latest document of ancient music that survives with ancient notation. It can hardly be doubted that elements of this tradition survived to influence the Gregorian music and other manifestations of musical performance that were subsequently to arise, and which have long been heard as underlying the development of Western music into the Renaissance and since.

It remains to use these new insights to produce plausible reconstructions and performances of the few dozen remaining Greek musical fragments, as I propose to do over the next few years; and to hope that more documents of ancient music will eventually come to light in museum collections and in archaeological discoveries. In the meantime, scholars are now encouraged, and enabled, to pursue in greater detail a fascinating question for which evidence is now available: what difference did music make to the poetic texts of ancient Greece?

Armand D’Angour and Tom Phillips have just edited and published Music Text and Culture in Ancient Greece (Oxford University Press)—a collection of essays from leading experts on the issues raised in this paper.