12 ms·

How did ancient Greek music sound?

106 points by sebkomianos 13y ago

corin_ 13y ago
> But isn't the music lost beyond recovery? The answer is no. The rhythms - perhaps the most important aspect of music - are preserved in the words themselves, in the patterns of long and short syllables. This sounds a little bit like bollocks to me. The way we would speak lyrics of more modern music, from Bach to Bieber, rarely matches up with how they are set to music. Given the article is written by the man behind this work, I'm not too sure whether that's a gross over-simplification for the BBC audience, or if it's simply the best they're able to do. Or perhaps some other research suggests that in those times singing really was identical to speaking with the exception of pitch/timbre etc. and rhythm didn't alter at all? edit: Does seem like it was just an over-simplification, some great additional thoughts in replies to me
- bagosm 13y ago
  Actually ancient Greek was a pretty melodic language. All words are ruled by mathematical rules, and there was clear rhythm. They had different letters for the same sound but different duration, such as Omega (ω) an Omikron (o) which means O-big and O-small. The only problem is that we don't really know the exact vocalization of all the various letters and notes (notes as combinations of letters, for example ΟΙ or ΕΙ) which have the same vocalization in new Greek. We do have however some clues as to how they were pronounced. All in all, because I have been following the subject for some time, we can't reproduce the music as it was. We have some directions as to how to rebuild the instruments and that's the best lead up to date, but still that's up for much debate as to how accurate the reconstruction is.
  - flurie 13y ago
    Actually, ω and o are not the same sound with different duration. You can refer to Vox Graeca pp. 60 & 71 for specifics.
    - xaritas 13y ago
      But close enough, if I understand, that an average Greek speaker might have accepted the statement as true, in the way that an average American English speaker might agree that "saw" and "pot" have the same vowel sound, no matter how many linguists' heads explode. I always wondered if the ω glyph developed when lazy monks sloppily doubled up oo to make the o-mega… but I never investigated whether this is true.
      - jeorgun 13y ago
        Completely off-topic, but I think linguists would be the first to point out that the two sounds are identical in many American English dialects. Also, from my understanding, ω is just a degenerate form of the upper-case form (Ω) when drawn really flat and with the edges flaired up [1]. [1] https://en.wikipedia.org/wiki/Omega https://en.wikipedia.org/wiki/Omega
      - bananaoomarang 13y ago
        I think ω is pronounced like 'or' and ο just pronounced 'o'
        bagosm 13y ago
        Replying to original comment as well as this. Ω is the capital of ω, exactly the same letters, though ancient Greeks only used capital letters. The Ω was pronounced a bit more deep in the throat, kind of like "ho", but that difference was more accented when the letter was used by itself as an article "Ω human", for example. Still though there is no absolute knowledge for the exact pronunciation of any of the ancient Greek. My belief is that the most close modern dialect is the Cypriot since it's the most melodic and rhythmic.
- asolove 13y ago
  In English, you can change the relative lengths of syllables in a word and it's still the same word. In Greek this isn't the case: the relative lengths of the vowels is part of what defines the word. Maybe you don't believe me, since syllable lengths are so fungible in English. But it really does sound wrong in classical Greek. A good way to get a feel for this is to imagine saying English words while putting the emPHAsis on the wrong syllaBLE. Sometimes it's just a bit weird, sometimes it's totally wrong, and sometimes it actually turns into a different word. You wouldn't write a poem and try to fit the meter by asking readers to change which syllable in a word is emphasized. (Or at least, you would do so only rarely and for a specific reason.) Now, in Greek, instead of emphasis, the relative length of vowels in syllables is prescribed and certain words are distinguished on this basis. In fact, whereas meter in English poetry is based on emphasized syllables, meter in classical Greek poetry is based on syllable length. So there is reason to think it wouldn't change just to fit a song melody. This is also why, though you can read Homer in English that is metrical, you can't get a feel for heroic verse without hearing it in the Greek. Let me give an indirect example that shows why this kind of reasoning doesn't work across languages. In English, we often indicate sarcasm, emphasis, or a question by intonation. A heavy, slow word might be more important, or a word that is pronounced with the tone rising might indicate a question or uncertainty. This is because "a word" in English is the same word regardless of the tone with which it is said. But in Chinese, the "same" sound with different tones actually is a different word. So in Chinese, you can't indicate a question with tone, because that would actually make a different word. Instead, you have to move words around, so that to indicate a word is the highlight of a sentence you move it to the beginning, or to indicate which word you aren't certain about you move it to the end. Human language is fascinating!
  - corin_ 13y ago
    Thanks for going into that - so in (ancient) Greek, were there words which were the same as each other but for syllable length? Your mention of syllable length being the basis of meter in poetry makes sense, but this is often the case in English too, while English music allows us poetic license to tweak away from the norm. I also think you're slightly wrong (or at least exaggerated) in saying that in English "you can't change which syllable is emphasized more than very rarely or it sounds like nonsense". If you take a piece of music and then alter this then sure, but actually the emphasis will often be slightly different to how it would be spoken. I wanted to double check on this belief, and thought about it while listening to Elton John's "I'm Still Standing" (which was already playing) and Stanford's Magnificat in G to add a different genre, in both cases the emphasis is often on different syllables to how we would speak it if we just saw the text without knowing the music. Anyway, thanks again for the reply, I guess it falls (at least to an extent) into the "simplified for the BBC audience", which is much preferable to the alternative of "it's bullshit". :) edit: Here's the Stanford piece (with music+lyrics if you're not familiar with the text already): http://www.youtube.com/watch?v=g7UorDdRg60 http://www.youtube.com/watch?v=g7UorDdRg60 Was just a random picking off the top of my head (added bonus that it's, in my opinion, one of the most beautiful things ever written)
- xaritas 13y ago
  This is music to accompany spoken poetry, so the it's not necessarily comparable to opera or pop music, where the voice is used as an instrument. In Greek lyric poetry the syllables are arranged according to fairly strict rules; having scanned (that is, marked up for meter) lots of Greek poetry, I don't doubt that this is an accurate way to get the rhythm of the piece. However, while I have no background in music theory, the claim that rhythm is the most important aspect to recovering the original sound doesn't seem right to me—I would assume that the tonality and harmonics are harder to create with certitude. Links of interest: http://en.wikipedia.org/wiki/Iambic_trimeter http://en.wikipedia.org/wiki/Iambic_trimeter and http://en.wikipedia.org/wiki/Trochaic_tetrameter http://en.wikipedia.org/wiki/Trochaic_tetrameter Interesting facts about ancient Greek: in contrast to English poetry, stress is indicated by the length of the vowel, rather than how forcefully it is pronounced; and vowels are accented by rising or falling pitch rather than stress. My professors said that ancient Greek probably sounded more like Chinese (which I gather follows similar rules) than a modern western language.
- EzGraphs 13y ago
  The timing in most modern music is metrical rather than driven by speech. Gregorian chant (or other forms of chant) are closer to what much ancient music was like in my understanding . This is also why a lot of poetry of the time is not characterized by huge amounts of metered syllabification and rhymes and the like (though some are cited in the article). I am basing this a bit on the idea that Jewish chant that later impacted the early church a few hundred years later is somewhat reflective of musical practice of the time.
  - corin_ 13y ago
    Is Gregorian chant based on speech (when it comes to rhythm)? Despite having sung a few pieces back in my musical days, this question never came into my head and I don't know nearly enough about Latin to make the comparison. It's late here in the UK so am about to sleep, but will Google around a bit tomorrow..
    - stan_rogers 13y ago
      Yes. It is also the case in plainchant in the C of E tradition (the "ancient office hymns" as well as the psalm, the Benedicite, the Sanctus and Benedictus, Te Deum etc.). It is only somewhat stylized so that it can be sung by groups of people (otherwise people's own idiosyncratic rhythms would produce a cacophony), and terminal syllables of verses tend to be held and elaborated, but it is very much based on the rhythms of ordinary speech.
    - EzGraphs 13y ago
      It is at least heavily weighted towards speech like the previous comment mentioned. This is difficult for modern folk to understand. In Anglican chant, the term "Anglican Thump" is used to describe a common problem where singers race in the initial section of the chant up to the cadence and then stop to pound out the last few notes in metered time.
- leoc 13y ago
  Well, at the minimum it's bollocks with an illustrious history. In the sixteenth century some people in Italy got the idea that the classical Greek plays must have been sung rather than spoken. Their attempts at reviving this probably-imaginary tradition are what gave us (European) opera. http://en.wikipedia.org/wiki/Florentine_Camerata http://en.wikipedia.org/wiki/Florentine_Camerata
  - christudor 13y ago
    Large parts of Classical Greek plays were were sung. We know this from a host of different sources (not least the plays themselves, parts of which are written in sung meters...)
    - leoc 13y ago
      Yes, but they weren't through-sung (were they?), which is what the Florentines seem to have decided they were (hence opera).
beeco 13y ago
Maybe they could do something like Fringe and get the noise from the walls.
Pxtl 13y ago
I wish they'd expored more the issue of the pitch itself. There are cultures on Earth that don't use the 12-semitone chromatic scale that is near-universal today. There are Eastern cultures that historically bore very different scales... to my ears it sounds unlistenable, but I'm always curious if that's just a cultural thing. Do we know the origin of the chromatic scale?
- mynameishere 13y ago
  Semitones fall so neatly into fractions that I can't imagine it's a matter of culture. When microtones are used I suspect tone deafness or intentional deviation/dissonance are the main actors.
  - twelvechairs 13y ago
    depends what you mean by 'semitones'. for instance, much of the worlds traditional music is based on pentatonic scales which have notes which are near (but not exactly on) the semitones of an equal tempered scale. generally the differences to the corresponding chromatic notes provide more consonance, not dissonance. 'microtones' is a funny word because it covers both consonant and dissonant sounds. there's some inherited cultural background to what is perceived as consonant too. a sitar may sound odd to us, and is 'microtonal' but that doesn't mean it is not musically consonant (notes are related in simple fractions)
  - hetman 13y ago
    I'm not sure if you can call the fractions of semi-tones of the equal tempered scale (used in most modern Western music) as neat. Indeed, its widespread introduction was very controversial because it was always considered a compromise on the purity of the sound. Never the less, its incredible convenience prevailed. When we talk about "pure" sounds. The sound intervals that sound best to the human ear form simple numeric frequency ratios. However, if you tune a scale in a way that is closer to this ideal, not every semi-tone is interchangeable and playing pieces in certain keys sounds down right horrible. That's why, choosing the key of a piece, once actually had a relatively larger impact on how it would sound, depending on how the instrument was tuned. Dissonance in some keys was even sometimes exploited by certain composers. Unfortunately we lose the impact of how these pieces would have sounded when played on modern instruments. We're talking here about pieces from composes as recent as Chopin.
davideous 13y ago
This reminds me of something my Latin teacher used to say: We don't know how Latin was actually pronounced because all of the audio tapes were destroyed when Rome burned down. :-)
hetman 13y ago
For anyone curious, here is a reconstruction of the Song of Seikilos mentioned in the article: http://www.youtube.com/watch?v=xERitvFYpAk http://www.youtube.com/watch?v=xERitvFYpAk It is based on the musical notation found on a tombstone and is, I think, the oldest complete Greek musical piece found to date.
- rednukleus 13y ago
  That is beautiful. I'm really looking forward to hearing more reconstructed ancient music.
lauremerlin 13y ago
Annie Belis, brilliant researcher from France's most selective post doct was already doing this with her ensemble Kerylos in 1990. The site is in French though but the songs in Ancient Greek http://www.kerylos.fr/index.php http://www.kerylos.fr/index.php And here is a chorus, easier for our ears I guess http://youtu.be/R_KmlIX3aHc http://youtu.be/R_KmlIX3aHc
christudor 13y ago
Here's my take on this: either the author or the BBC has simplified what is really going on, because we actually know a lot more about Ancient Greek Music than this article suggests. Most importantly (I think), ancient music is not (only) preserved in the words themselves (i.e. by vowel length and things like that). There is also a lot of evidence for musical notation, which has been preserved both in the papyrus tradition (where we have lines of tragic verse with the actual notes transcribed above - you can see an example here: http://classics.uc.edu/music/ http://classics.uc.edu/music/), as well as from discussions of ancient music by the ancients themselves, e.g. Plutarch's 'On Music'

106 points by sebkomianos 13y ago

> But isn't the music lost beyond recovery? The answer is no. The rhythms - perhaps the most important aspect of music - are preserved in the words themselves, in the patterns of long and short syllables. This sounds a little bit like bollocks to me. The way we would speak lyrics of more modern music, from Bach to Bieber, rarely matches up with how they are set to music. Given the article is written by the man behind this work, I'm not too sure whether that's a gross over-simplification for the BBC audience, or if it's simply the best they're able to do. Or perhaps some other research suggests that in those times singing really was identical to speaking with the exception of pitch/timbre etc. and rhythm didn't alter at all? edit: Does seem like it was just an over-simplification, some great additional thoughts in replies to me

bagosm 13y ago
Actually ancient Greek was a pretty melodic language. All words are ruled by mathematical rules, and there was clear rhythm. They had different letters for the same sound but different duration, such as Omega (ω) an Omikron (o) which means O-big and O-small. The only problem is that we don't really know the exact vocalization of all the various letters and notes (notes as combinations of letters, for example ΟΙ or ΕΙ) which have the same vocalization in new Greek. We do have however some clues as to how they were pronounced. All in all, because I have been following the subject for some time, we can't reproduce the music as it was. We have some directions as to how to rebuild the instruments and that's the best lead up to date, but still that's up for much debate as to how accurate the reconstruction is.
- flurie 13y ago
  Actually, ω and o are not the same sound with different duration. You can refer to Vox Graeca pp. 60 & 71 for specifics.
  - xaritas 13y ago
    But close enough, if I understand, that an average Greek speaker might have accepted the statement as true, in the way that an average American English speaker might agree that "saw" and "pot" have the same vowel sound, no matter how many linguists' heads explode. I always wondered if the ω glyph developed when lazy monks sloppily doubled up oo to make the o-mega… but I never investigated whether this is true.
    - jeorgun 13y ago
      Completely off-topic, but I think linguists would be the first to point out that the two sounds are identical in many American English dialects. Also, from my understanding, ω is just a degenerate form of the upper-case form (Ω) when drawn really flat and with the edges flaired up [1]. [1] https://en.wikipedia.org/wiki/Omega https://en.wikipedia.org/wiki/Omega
    - bananaoomarang 13y ago
      I think ω is pronounced like 'or' and ο just pronounced 'o'
      - bagosm 13y ago
        Replying to original comment as well as this. Ω is the capital of ω, exactly the same letters, though ancient Greeks only used capital letters. The Ω was pronounced a bit more deep in the throat, kind of like "ho", but that difference was more accented when the letter was used by itself as an article "Ω human", for example. Still though there is no absolute knowledge for the exact pronunciation of any of the ancient Greek. My belief is that the most close modern dialect is the Cypriot since it's the most melodic and rhythmic.
asolove 13y ago
In English, you can change the relative lengths of syllables in a word and it's still the same word. In Greek this isn't the case: the relative lengths of the vowels is part of what defines the word. Maybe you don't believe me, since syllable lengths are so fungible in English. But it really does sound wrong in classical Greek. A good way to get a feel for this is to imagine saying English words while putting the emPHAsis on the wrong syllaBLE. Sometimes it's just a bit weird, sometimes it's totally wrong, and sometimes it actually turns into a different word. You wouldn't write a poem and try to fit the meter by asking readers to change which syllable in a word is emphasized. (Or at least, you would do so only rarely and for a specific reason.) Now, in Greek, instead of emphasis, the relative length of vowels in syllables is prescribed and certain words are distinguished on this basis. In fact, whereas meter in English poetry is based on emphasized syllables, meter in classical Greek poetry is based on syllable length. So there is reason to think it wouldn't change just to fit a song melody. This is also why, though you can read Homer in English that is metrical, you can't get a feel for heroic verse without hearing it in the Greek. Let me give an indirect example that shows why this kind of reasoning doesn't work across languages. In English, we often indicate sarcasm, emphasis, or a question by intonation. A heavy, slow word might be more important, or a word that is pronounced with the tone rising might indicate a question or uncertainty. This is because "a word" in English is the same word regardless of the tone with which it is said. But in Chinese, the "same" sound with different tones actually is a different word. So in Chinese, you can't indicate a question with tone, because that would actually make a different word. Instead, you have to move words around, so that to indicate a word is the highlight of a sentence you move it to the beginning, or to indicate which word you aren't certain about you move it to the end. Human language is fascinating!
- corin_ 13y ago
  Thanks for going into that - so in (ancient) Greek, were there words which were the same as each other but for syllable length? Your mention of syllable length being the basis of meter in poetry makes sense, but this is often the case in English too, while English music allows us poetic license to tweak away from the norm. I also think you're slightly wrong (or at least exaggerated) in saying that in English "you can't change which syllable is emphasized more than very rarely or it sounds like nonsense". If you take a piece of music and then alter this then sure, but actually the emphasis will often be slightly different to how it would be spoken. I wanted to double check on this belief, and thought about it while listening to Elton John's "I'm Still Standing" (which was already playing) and Stanford's Magnificat in G to add a different genre, in both cases the emphasis is often on different syllables to how we would speak it if we just saw the text without knowing the music. Anyway, thanks again for the reply, I guess it falls (at least to an extent) into the "simplified for the BBC audience", which is much preferable to the alternative of "it's bullshit". :) edit: Here's the Stanford piece (with music+lyrics if you're not familiar with the text already): http://www.youtube.com/watch?v=g7UorDdRg60 http://www.youtube.com/watch?v=g7UorDdRg60 Was just a random picking off the top of my head (added bonus that it's, in my opinion, one of the most beautiful things ever written)
xaritas 13y ago
This is music to accompany spoken poetry, so the it's not necessarily comparable to opera or pop music, where the voice is used as an instrument. In Greek lyric poetry the syllables are arranged according to fairly strict rules; having scanned (that is, marked up for meter) lots of Greek poetry, I don't doubt that this is an accurate way to get the rhythm of the piece. However, while I have no background in music theory, the claim that rhythm is the most important aspect to recovering the original sound doesn't seem right to me—I would assume that the tonality and harmonics are harder to create with certitude. Links of interest: http://en.wikipedia.org/wiki/Iambic_trimeter http://en.wikipedia.org/wiki/Iambic_trimeter and http://en.wikipedia.org/wiki/Trochaic_tetrameter http://en.wikipedia.org/wiki/Trochaic_tetrameter Interesting facts about ancient Greek: in contrast to English poetry, stress is indicated by the length of the vowel, rather than how forcefully it is pronounced; and vowels are accented by rising or falling pitch rather than stress. My professors said that ancient Greek probably sounded more like Chinese (which I gather follows similar rules) than a modern western language.
EzGraphs 13y ago
The timing in most modern music is metrical rather than driven by speech. Gregorian chant (or other forms of chant) are closer to what much ancient music was like in my understanding . This is also why a lot of poetry of the time is not characterized by huge amounts of metered syllabification and rhymes and the like (though some are cited in the article). I am basing this a bit on the idea that Jewish chant that later impacted the early church a few hundred years later is somewhat reflective of musical practice of the time.
- corin_ 13y ago
  Is Gregorian chant based on speech (when it comes to rhythm)? Despite having sung a few pieces back in my musical days, this question never came into my head and I don't know nearly enough about Latin to make the comparison. It's late here in the UK so am about to sleep, but will Google around a bit tomorrow..
  - stan_rogers 13y ago
    Yes. It is also the case in plainchant in the C of E tradition (the "ancient office hymns" as well as the psalm, the Benedicite, the Sanctus and Benedictus, Te Deum etc.). It is only somewhat stylized so that it can be sung by groups of people (otherwise people's own idiosyncratic rhythms would produce a cacophony), and terminal syllables of verses tend to be held and elaborated, but it is very much based on the rhythms of ordinary speech.
  - EzGraphs 13y ago
    It is at least heavily weighted towards speech like the previous comment mentioned. This is difficult for modern folk to understand. In Anglican chant, the term "Anglican Thump" is used to describe a common problem where singers race in the initial section of the chant up to the cadence and then stop to pound out the last few notes in metered time.
leoc 13y ago
Well, at the minimum it's bollocks with an illustrious history. In the sixteenth century some people in Italy got the idea that the classical Greek plays must have been sung rather than spoken. Their attempts at reviving this probably-imaginary tradition are what gave us (European) opera. http://en.wikipedia.org/wiki/Florentine_Camerata http://en.wikipedia.org/wiki/Florentine_Camerata
- christudor 13y ago
  Large parts of Classical Greek plays were were sung. We know this from a host of different sources (not least the plays themselves, parts of which are written in sung meters...)
  - leoc 13y ago
    Yes, but they weren't through-sung (were they?), which is what the Florentines seem to have decided they were (hence opera).

beeco 13y ago

Maybe they could do something like Fringe and get the noise from the walls.

Pxtl 13y ago

I wish they'd expored more the issue of the pitch itself. There are cultures on Earth that don't use the 12-semitone chromatic scale that is near-universal today. There are Eastern cultures that historically bore very different scales... to my ears it sounds unlistenable, but I'm always curious if that's just a cultural thing. Do we know the origin of the chromatic scale?

mynameishere 13y ago
Semitones fall so neatly into fractions that I can't imagine it's a matter of culture. When microtones are used I suspect tone deafness or intentional deviation/dissonance are the main actors.
- twelvechairs 13y ago
  depends what you mean by 'semitones'. for instance, much of the worlds traditional music is based on pentatonic scales which have notes which are near (but not exactly on) the semitones of an equal tempered scale. generally the differences to the corresponding chromatic notes provide more consonance, not dissonance. 'microtones' is a funny word because it covers both consonant and dissonant sounds. there's some inherited cultural background to what is perceived as consonant too. a sitar may sound odd to us, and is 'microtonal' but that doesn't mean it is not musically consonant (notes are related in simple fractions)
- hetman 13y ago
  I'm not sure if you can call the fractions of semi-tones of the equal tempered scale (used in most modern Western music) as neat. Indeed, its widespread introduction was very controversial because it was always considered a compromise on the purity of the sound. Never the less, its incredible convenience prevailed. When we talk about "pure" sounds. The sound intervals that sound best to the human ear form simple numeric frequency ratios. However, if you tune a scale in a way that is closer to this ideal, not every semi-tone is interchangeable and playing pieces in certain keys sounds down right horrible. That's why, choosing the key of a piece, once actually had a relatively larger impact on how it would sound, depending on how the instrument was tuned. Dissonance in some keys was even sometimes exploited by certain composers. Unfortunately we lose the impact of how these pieces would have sounded when played on modern instruments. We're talking here about pieces from composes as recent as Chopin.

davideous 13y ago

This reminds me of something my Latin teacher used to say: We don't know how Latin was actually pronounced because all of the audio tapes were destroyed when Rome burned down. :-)

hetman 13y ago

For anyone curious, here is a reconstruction of the Song of Seikilos mentioned in the article: http://www.youtube.com/watch?v=xERitvFYpAk http://www.youtube.com/watch?v=xERitvFYpAk It is based on the musical notation found on a tombstone and is, I think, the oldest complete Greek musical piece found to date.

rednukleus 13y ago
That is beautiful. I'm really looking forward to hearing more reconstructed ancient music.

lauremerlin 13y ago

Annie Belis, brilliant researcher from France's most selective post doct was already doing this with her ensemble Kerylos in 1990. The site is in French though but the songs in Ancient Greek http://www.kerylos.fr/index.php http://www.kerylos.fr/index.php And here is a chorus, easier for our ears I guess http://youtu.be/R_KmlIX3aHc http://youtu.be/R_KmlIX3aHc

christudor 13y ago

Here's my take on this: either the author or the BBC has simplified what is really going on, because we actually know a lot more about Ancient Greek Music than this article suggests. Most importantly (I think), ancient music is not (only) preserved in the words themselves (i.e. by vowel length and things like that). There is also a lot of evidence for musical notation, which has been preserved both in the papyrus tradition (where we have lines of tragic verse with the actual notes transcribed above - you can see an example here: http://classics.uc.edu/music/ http://classics.uc.edu/music/), as well as from discussions of ancient music by the ancients themselves, e.g. Plutarch's 'On Music'