Accurate Greek

[WilliamThomson]

Does anyone know if this is representative of how ancient Greek sounded?

http://www.rhapsodes.fll.vt.edu/Greek.htm

Thanks in advance.

WT

[Qimmik]

No one alive today can tell you definitively whether this is how ancient Greek sounded. It's a plausible guess by someone who has presumably studied all there is to know about ancient Greek phonetics, but the ancient Greeks hadn't developed a notation to record the sounds of ancient Greek at a more fine-grained level than the actual letters of the alphabet--certainly not anything like the IPA.

It's also true that some of the sounds of even the purest Attic, spoken, say, from 480 to 323 BC, must have shifted somewhat over that period, just as the pronunciation of standard American or British English has changed over much shorter periods. And there must have been many local and class-based variations of Attic, without even going into the bewildering profusion of dialects prevalent throughout the Greek world.

[Scribo]

No it's awful and so strongly Americophonic. I can see where Daitz is trying to draw from leading research but he's certainly not articulating anything spoken in the Mediterranean - outside of American tourists, Germanic condottieri etc.

[Qimmik]

For one thing, I suspect that differences of vowel quantity are exaggerated in the clip. Without a substantial amount of practice, speakers of languages like English without distinctive vowel quantity would probably have difficulty hearing differences in vowel quantity, at least in normal conversation, except perhaps for long vowels with a circumflex accent. (But distinguishing Greek pitch accents would probably take some practice, too.) And quite possibly there was a slight difference in quality between short vowels and the corresponding long vowels. To native speakers, differences in vowel quantity would not be a problem.

Here is a You-Tube clip with some samples of Czech, a language with distinctive vowel quantity. Vowel length is indicated by an acute accent (or, in the case of u, by a little circle above the letter). There are several repetitions of each phrase, with an increasing rapid tempo--the last repetition seems to reflect everyday speech. Although long vowels are stretched out in the initial repetitions, I suspect that in the most conversational repetitions, most English speakers, coming at this cold, would have considerable difficulty telling which vowels are long and which are short, where native speakers of Czech would have no trouble. (I'm assuming that the individuals speaking in the clip are native speakers.) I wonder whether the same wouldn't be true in ancient Greek or classical Latin.

http://www.youtube.com/watch?v=bY9cp7r2CZc

And here's a clip with a number of people speaking Hungarian, another language with distinctive vowel quantity. There's no English or written Hungarian guidance. (I have no idea what this is about--it seems to be a news broadcast.) Trying listening for vowel quantity (and don't be deceived by the strong Hungarian word-initial stress accent).

http://www.youtube.com/watch?v=9N2oviuhMVc

Perhaps vowel quantity was exaggerated in reading or reciting ancient Greek poetry, but who knows?

[Scribo]

100% its more subtle than people thing, or rather if you're not used to it. Obviously recitation of (poetic) texts was different, more...artful I guess, as we can gleam from various references strewn throughout antiquity. But as for everyday speech, I recently brought up Punjabi which employs both long/short distinctions and tone and though the latter is important for distinguishing meaning its hardly blatant. I recently got to listen to some Virah/Vaarah (lit: heroic) poetry, high metrical descendent verse form from the more complex Sanskrit lyrics. One could hear the tones and quantity at work somewhat more than spoken speech. But then I'm used to such things and tend to be good with accents.

We are getting a lot of topics on oral stuff lately, perhaps somewhat ought to collate a mega thread.

[Qimmik]

It's worth taking a look at these Wikipedia articles on Swedish and Norwegian phonology. Like ancient Greek, Swedish and Norwegian have two tone accents. The article gives some idea of the complexity of the Swedish and Norwegian situations, and we have no way of knowing in any detail what the situation might have been like in ancient Greek.

http://en.wikipedia.org/wiki/Swedish_phonology#Stress_and_pitch

http://en.wikipedia.org/wiki/Norwegian_phonology#Accent

It's also worth bearing in mind that for the most part, the diacritical marks for the pitch accents in ancient Greek weren't invented or assigned to words until the period when the language was shifting to a stress-based accent--the diacritical marks were supposedly intended to help readers who may not have been fully used to the tonal accent system.

[pster]

No it's awful and so strongly Americophonic. I can see where Daitz is trying to draw from leading research but he's certainly not articulating anything spoken in the Mediterranean - outside of American tourists, Germanic condottieri etc.

You more or less singlehandedly turned me off to Diatz! I guess discovering Hagel didn't hurt. But I think the problem with Daitz is that Greek couldn't have sounded like that because no language sounds like that. No language sounds that exaggerated. Even ones that seem a bit exagerated like Texan English or Québecois French have more of a droning twang than a continual bobbing and weaving where every third syllable seems emotionally fraught.

Daitz does however have the largest set of recordings.

[Scribo]

You more or less singlehandedly turned me off to Diatz!
...
Daitz does however have the largest set of recordings.

...and the heart within me was pleased. Yes its problematic that he's the largest single set, but there are others too and I think its better to have a little good than a big bad. Wow. today I'm basically Kallimachos.

pros ton Qimmik:

Indeed, I think they're also good data sets but I'm not too familiar with them, though some of my paternal cousins grew up in Norway and thus I've heard Norwegian a lot so it ought not to be too hard a jump. Have you seen Stephen and Devine's "Prosody of Greek Speech?" If not I can try and re-read and synthesise some of the stuff together but its nigh on 600 pages. But it has a lot of good stuff.

[demetri]

No it's awful and so strongly Americophonic. I can see where Daitz is trying to draw from leading research but he's certainly not articulating anything spoken in the Mediterranean - outside of American tourists, Germanic condottieri etc.

You more or less singlehandedly turned me off to Diatz! I guess discovering Hagel didn't hurt. But I think the problem with Daitz is that Greek couldn't have sounded like that because no language sounds like that. No language sounds that exaggerated. Even ones that seem a bit exagerated like Texan English or Québecois French have more of a droning twang than a continual bobbing and weaving where every third syllable seems emotionally fraught.

Daitz does however have the largest set of recordings.

Chuckle...I had to wash my ears out after listening to Daitz.

[Markos]

pster wrote: But I think the problem with Daitz is that Greek couldn't have sounded like that because no language sounds like that. No language sounds that exaggerated.

When I played Daitz recently, my twelve-year-old over heard it and literally burst into laughter. I say this not to be unkind--I have respect and sympathy for Daitz' efforts--but to point out that pster is probably correct. Evan der Millner recently posted some videos where he too tried to accurately reproduce the tones

http://www.youtube.com/watch?v=e-TnIMunLBY

and I think these were not well received because people thought he also sounded overly exaggerated. I don't really have a problem with either Daitz or Evan--I've never met a profora I did not like--but the exaggerated tones are at the very least a distraction.

It's a balance. You try to be historically accurate but as a practical matter the tones come out sounding funny. So, what do you do? Note that the JACT folks, who made an audio in which they too seek historically accuracy, decided to simply not reproduce the tones. I know about thirty people who actively speak Ancient Greek as living language, and all of them more or less ignore the tones, although Rico usually does something a little tonal with the circumflex.

Certain accomplished Greek speakers--Randall Buth and Roberto Lionello come to mind--solve this problem I think, but giving their diction a slightly tonal, musical sound, without attempting to reproduce the specific tones and pitches.

You've got a tough decision to make, William. If you replace the tonal accents with simple stress accents in your Homeric flashcard project, you will probably not be historically accurate, but this is what I think you should do.

[Paul Derouda]

My point is not to mock Daitz, but rather I think with different linguistic backgrounds we can teach each other by showing stuff that might be difficult for others.

My native language is Finnish where vowel quantity is very important and to me Daitz seems to exaggerate vowel quantity a LOT. Though of course it's theoretically possible maybe that a Greek long vowel was longer than a Finnish one... Another thing he does is that he pronounces short diphthongs like they were long, so his long diphthongs are really really REEEEEAAAALLY long.

I listened to him and noted here the specific instances with a "short" diphthong that in my opinion could be taken as model on how to pronounce a LONG dipthong:

ευ, υι, ἀθροίζω, ἐντεῦθεν, Άθηναῖος.

Άθηναῖος is especially clear and this shows exactly how I think you should pronounce long alpha subscript /a:i/ (ᾳ)! His ἐντεῦθεν is maybe more like /eu:/ than /e:u/.

Disclaimer: Finnish doesn't have long diphtongs. But still I think I have an idea.

What do you think about this, especially the pitch accent:
http://prosoidia.com/agamemnon-of-aeschylus/

I think it's pretty good, the only "problem" I heard that φ θ χ are treated as fricatives (English f, th, h), not aspirates. The long vowels seem just about right to me. In another thread we were thinking with Stirling the speaker may been Greek.

[Zetes]

It's worth taking a look at these Wikipedia articles on Swedish and Norwegian phonology. Like ancient Greek, Swedish and Norwegian have two tone accents. The article gives some idea of the complexity of the Swedish and Norwegian situations, and we have no way of knowing in any detail what the situation might have been like in ancient Greek.

It's also worth bearing in mind that for the most part, the diacritical marks for the pitch accents in ancient Greek weren't invented or assigned to words until the period when the language was shifting to a stress-based accent--the diacritical marks were supposedly intended to help readers who may not have been fully used to the tonal accent system.

I've recently asked for some advice on another aspect of pronunciation (aspiration), but I must admit this tone/pitch accent subject is something I haven't troubled about; this is partly because I have spent time ( 15 months or so) living in Sweden and was able to understand and make myself understood without bothering about 1st and 2nd tones at all. Swedish is still quite comprehensible without them - not good or really correct - but easily enough understood.

[Fritz]

Unfortunately, the above-mentioned site http://prosoidia.com/ seems to have been hacked by someone called "Dowoh." Strange target. Thus I can't utilize it for help with my vowel problem, which concerns eta.

I've read Allen, listened to Daitz, as well as some others, and still can't seem to get a clear sense of how eta ought to be pronounced. Of all the vowels this one has given me the most persistent trouble.

In most recordings, to my ears it does not sound like /ɛ:/ or the e in French tête when Daitz and others pronounce, for example, Mῆνιν. It seems to shade more open and central. Which creates difficulty for me as it appears to approach alpha. And my habitual, unintentional tendency is to add a glide and force it closer to the diphthong ει

I'd appreciate any insight or leads to further information, posts, etc.

This forum is an amazing resource, so many thanks!

F.

[Qimmik]

Three reasons why modern attempts to recapture with precision the pronunciation of ancient Greek texts are futile:

(1) The ancient Greek alphabet, although very well-adapted to the language, was not an exact phonetic transcription.

(2) Our access to ancient Greek is largely through texts that have complicated and not entirely understood histories (though epigraphic evidence sheds some light).

(3) The ancient Greek "language" encompassed many dialects and Greek has been constantly changing over the entire course of its history.

[Qimmik]

An example of how the texts at our disposal have undergone revisions that obscure the original pronunciation.

Alcman's Partheneion, thought to have been written in the seventh century BCE, is preserved in a papyrus now in the Louvre that is believed to date from the 1st century CE. The text includes the famous line: εcτι τιc cιων τιcιc -- "there is a certain vengeance of the gods." The form cιων is Laconian for θεῶν. In Spartan speech, the sound represented by theta at some point underwent a change to a sound represented by sigma (which may have been a crude way of representing a dental fricative like the unvoiced th sound in English--a change that eventually occurred in all Greek by the imperial period).

But the epigraphic evidence indicates that this change didn't occur in Laconian until sometime around 400 BCE--a couple of centuries after Alcman's time. If that's the case, Alcman must have written θιων or θεων, and the form cιων in the Louvre papyrus must have been substituted sometime after the phonetic change occurred by someone who was unaware of the date of the change and, based on contemporary Spartan speech, thought that Alcman would have written cιων rather than θιων or θεων.

[Albannach]

Both my wife (a native speaker of Scottish Gaelic) and I (native speaker of English, fluent in German and Gaelic too, currently learning a tonal language, Mandarin Chinese) thought that Daitz' voice was too forced.

This sounded much easier on the ear:

http://www.oeaw.ac.at/kal/agp/

Here's another interesting one, but it seems to omit the use of tones for stress:

http://www.youtube.com/watch?v=lUYWSY0yCmk

To go in the other direction, this is ostensibly Attic Greek, but is instantly recognisable as pronunciation from speakers of US English:

http://ucbclassics.dreamhosters.com/anc ... lphaU.html

Ouch!

There has been some work done using the other clues to pronunciation, the musical settings (including the choice of instrument), which help to determine the parameters of pronunciation:

http://www.oeaw.ac.at/kal/sh/

To me, this sounds the most convincing recording that I've tracked down so far. (It reminds me to some extent of the chants/songs from the Ulster and Fenian lays, which tend to be in very archaic language, often including "runs" that seem to represent older, ossified forms of language - just as in Homer.)

Well done to Daitz for trying to produce something, but can perhaps a speaker of modern Greek used to tonal languages offer a superior version, based on the philology that has been done?

It's very frustrating that no convincing attempt seems to have been made available online, although there is surely sufficient information available to indicate the major part of how Homer was pronounced. (Despite changes in normal pronunciation, recitations still took place over a long period; perhaps some modernisation towards Attic Greek was usual in most productions in the past.)

Here's a comment from a German-language broadcaster on the solution to pronouncing Homer: "Homer müsste man eigentlich singen: Jeder Hexameter hat seinen eigenen Melodiebogen, der auf dem musikalischen Akzent der einzelnen Wörter aufbaut." ("Homer really has to be sung: each hexameter has its own musical phrase which is based on the tones of individual words.")

In the end, it seems that despite the way classicists have embraced the internet, there is not enough confidence to propose a range of solutions for those, like myself and the OP, that would like to learn Homeric Greek with a fair degree of certitude that people pronounced it roughly in the way we are rendering it.

[pakmunsu]

Athenaze seems to use the pitch accent system, and claims to follow Daitz, without sounding as weird and forced as him.

http://www.youtube.com/watch?v=A77JzJN4W4g

I'm thinking of buying Athenaze. As a side note, the Italian version of Athenaze is reputed to be written more along the lines of the Orberg natural method of Lingua Latina.

[jeidsath]

Well done to Daitz for trying to produce something, but can perhaps a speaker of modern Greek used to tonal languages offer a superior version, based on the philology that has been done?

I can't help you there, but I did read Vox Graeca today, and then spent 30 minutes watching Aladdin in Norwegian trying to pick up their pitch pattern. I hereby present what may be the world's first "Fargo-inspired" ancient Greek dialogue:

http://youtu.be/h3HngJINGhI

Having worked on pitch for a number of weeks now, I have to state that it is *far* easier to read a text when there is whole-word involvement and a bit of rhythm. Also, the rising accent at the end of sentences (as in Norwegian) seems to really fit the Greek.

[Cheiromancer]

Having worked on pitch for a number of weeks now, I have to state that it is *far* easier to read a text when there is whole-word involvement and a bit of rhythm. Also, the rising accent at the end of sentences (as in Norwegian) seems to really fit the Greek.

I'm reminded of a post by annis who writes that

The acute was not an across-the-board rising accent. In most of the world's pitch accent system, the pitch accent is not identified by being at a higher pitch than previous syllables, but because the following syllable is at a lower pitch. So, using 1-9 as a rough pitch guide,

ἄνθρωπός τις 545 4 (possibly 3 for 4)

Only on a long vowel or diphthong would the acute represent what sounds like a rising pitch. And the circumflex was very probably just a falling pitch, not a rise then fall. The grave was merely the absence of a pitch accent — definitely not a falling contour in any interpretation of the evidence.

I periodically try to improve my pronunciation of Greek, but there are just so many issues. How the vowels work (especially eta and omega). Hearing the aspiration of consonants (theta vs tau, etc.) Vowel quantity. Where stress falls if it is decoupled from tone. It is too hard for me to untangle, and I don't know of any verbal models (like your YouTube videos) that I really trust.

Still, I'm glad that some people are still working on the problem, and putting things out there for critique. Kudos.

I wonder if a computer could be programmed to read Ancient Greek aloud according to various proposed rules? If a serviceable model could be discovered, learning to speak Ancient Greek would be much easier. You'd just have to mimic the machine. There would probably be several such models, but that's fine; there was a great range of dialects in Ancient Greek. But even one reliable model to base one's studies on would be incredibly valuable.

[jeidsath]

It is too hard for me to untangle, and I don't know of any verbal models (like your YouTube videos) that I really trust.

I am a brand-new Greek learner, so definitely don't take my videos as anything but experiments! But I can only think of a few reasons to use an ancient Greek pronunciation rather than a modern Greek pronunciation:

1. Vowel quantity is wrong in modern Greek, and this breaks poetry
2. Many vowels are no longer distinct in modern Greek, and this means you can't spell what you've heard, and a lot of homonyms begin to appear.
3. Modern Greek has lost its tonal structure, and this (along with 2) breaks poetry and reduces the euphony of the language.

Other differences are far less important, to my mind. I am working on r-trilling at the moment (it's hit and miss right now). But that kind of thing is much less important than the above 3 points.

My guess is that if you spent enough time in conversational ancient Greek, working on vowel quantity, vowel sound, and tone, you'd hit the good enough stage. You wouldn't sound like a native of Attica from 400 BC., but you'd be close enough for all practical purposes, including language euphony.