The story of Heinrich Schliemann hearing the drunken miller recite Homer, inspiring him to learn Greek and go on to dig up Troy, has to have a certain resonance to any modern-day reciter of ancient Greek. A certain intoxication, even without drink, is familiar to many who have attempted it, a feeling that, with tongue and breath rather than shovels, we are ourselves digging up ancient treasures. Read well or badly, it is clear that this is poetry designed to be heard. And ideally, we would like to be like the miller in capturing the interest of people who do not yet know the language. Though naturally, it would be a bit disappointing to find only such, and not even the occasional Greek scholar, in our audience.
The ancient sounds represented by the Greek alphabet are actually fairly well known. One could argue, in any case, that exact reproduction of the ancient phonemes is not even what is most crucial to a good pronunciation. They varied a good deal anyway over different regions, periods, and dialects, just as English does. The real problems arise in the area of rhythm and prosody.
As regards rhythm, three ways of reading Greek aloud are currently taught in American colleges. Most students learn at least two of them. All are inconsistent with one another, and all are just different ways of imposing our native speech rhythms on a language they don’t really fit.
The first way puts the emphasis on the syllables that carry accent marks in the texts. Now in English, when we accent a syllable, we generally either speak it at a higher pitch, or prolong it, or both. But in Greek it was done by pitch alone, and a syllable was prolonged only if it contained a long vowel, or if the vowel was followed by multiple consonants. A long syllable could be accented or not, and so could a short one. The two things were quite independent of one another. Speaking Greek with an English-style accentuation (a type of “stress accent,” which many other languages, including modern Greek, have) results in many syllables being either longer or shorter than they ought to be.
That’s the method used for prose. For verse, where the pattern of long and short syllables is the basis of the meter, a second method of reading has to be learned. This one ignores the accent marks, and puts the emphasis on the long syllables. Verses will now scan correctly, after a fashion, though the rhythm is still off, because our native speech habits insist on treating pitch and duration as more or less interchangeable. Some or all of the “long” syllables will therefore come out short, though spoken at a higher pitch (whether accented in the text or not), while accents that fall on short syllables will not be observed at all.
For reading verse, therefore, a third way of reading is often taught, in which every long syllable is deliberately made twice as long as a short one, according to the ratio theoretically observed in all Greek verse. This represents the correct rhythm a bit more adequately. But the native English-speaker (and speakers of other stress-accented languages), hearing the prolonged syllables as accented, will mostly try to put the pitch emphasis on them as well. All accented short syllables and many unaccented long ones will therefore be given the wrong intonation. And overall the rhythm, with the same syllables carrying both kinds of emphasis, will be very monotonous.
Why are students not taught from the start to keep accent and syllable length separate? Because it’s just too difficult, is the usual response, and because we have better things to do with our time. We don’t really know how it sounded. Why teach students something that will certainly sound outlandish and be difficult to remember, with no assurance that it is even correct?
In poetry, however, especially ancient Greek poetry, rhythm is more than a trivial side issue. By long-standing tradition, a passion for deep metric analysis has often gone hand in hand with total lack of interest in the proper rendition of rhythm in performance. What sort of “understanding” of the rhythms in question this really represents, I leave to the scholarly world to answer.
As a non-scholar, concerned mostly with finding a way of reading Greek poetry rhythmically and expressively for my own satisfaction, I make no great claims for the “authenticity” of my pronunciation. For many aspects of it, I have no more scholarly argument than the fact that to me it sounds better that way. I think we could learn here from the experience of the early music movement of the last few decades. In the beginning the performances often suffered from pedantry: it didn’t matter how bad it sounded, as long as it could be proved to be “authentic.” By now, however, it is pretty much agreed that “authenticity” is a chimera; most early music performers nowadays, if they are any good, think of themselves as artists first and scholars second. This doesn’t mean that scholarship is abandoned, only that aesthetic arguments for particular practices are once more paramount.
For particular aspects of Greek pronunciation and verse recitation, a few of my own arguments, both aesthetic and quasi-scholarly, can be found in my remarks following the readings, and in the notes for classicists. Mostly, though, it is the performances themselves that will either convince, or not.