I just checked my Langenscheidt–sad day when you have to go to Germans to learn how to pronounce French!–and they indicate ɛ: for même but ɛ for très. I’ve always tried to be aware of the distinctions between ɛ:, ɛ and e, at least since my university days anyway. My professor was the head of the department, was British, and had all the personality of a steel wool pad, but had learned all her French in a vacuum (or so the story went) and so was a hyper-stickler for detail, and I still remember her making a big deal about how to pronounce tête. So after that class I made a point of mastering the phonetic alphabet in my dictionary. The next year Francophones told me that I sounded very Parisian. 
Yes, Brel’s ê is particularly open to my ear. Maybe that doesn’t mean anything. Maybe I’m just voicing my own uncertainty about French e and ɛ like you’re voicing yours about Greek ει and η. Anyway, I think the problems are analogous.
As far as I know, difference between vowel quantity is never phonemical in French, so I wouldn’t worry about ɛ: and ɛ. I think quantity changes according to position in the clause, emphasis etc. I’m not so sure though, it comes naturally to me, since I’ve never really read those rules in a book.
And yeah, I can imagine how it makes you sound like a Parisian. Pronounce all those final e’s and you’ll sound like you’re from Marseilles…
Btw, Markos’ suggestion sounded pretty good to me.
Aren’t there two issues that you have put on the table?
ɛ: vs. ɛ in Brel’s même and today’s même,
ɛ vs. e in allait and allé.
I’d like to pretend I know what you mean, but I don’t. Can you explain this? Do you mean that there is no utterance whose meaning would change if we changed a short vowel to a long one or vice versa?
I don’t like Maaarkos’ suggestion any more than Scriiibo’s. a:/a and i:/i are already oversubscribed.
il vs. île is i vs i:
Wouldn’t that difference in quantity be phonemical?
My dictionary says gives il [il] and île [il], même [mɛm] and très [trɛ]. I would usually pronounce all these vowels short. I could also pronounce all of these long in some contexts, probably even il, I can’t say exactly when but I think it has to do with emphasis or clause position or something.
I had another look at the Wikipedia article: “With the exception of the distinction made by some speakers between /ɛ/ and /ɛː/ in rare minimal pairs like mettre [mɛtʁ] (‘to put’) vs. maître [mɛːtʁ] (‘teacher’), variation in vowel length is entirely allophonic.” So maybe that’s part of what surprises me about Brel’s ê, though I still suspect his ɛ is more open than mine…
So if your sources say [mɛ:m], maybe that reflects “some speakers’s” way of speaking which some other people consider standard.
Argh, I don’t know, haven’t ever properly studied this stuff except by speaking.
Markos’ suggestion is an ok approximation in my opinion if we don’t seem to be able to nail down [ɛ:] and [e:] properly with our linguistic backgrounds.
It seems like your dictionary doesn’t use the colon at all. Am I right about that? The online French dictionary I use is like that. My understanding is that the phonetic alphabet rarely fully deployed and what one gets is some simplification depending on who the audience is. I love precision and tradition, but if my dictionary used the full alphabet, I’d change my ways.
https://upload.wikimedia.org/wikipedia/commons/1/15/IPA_chart_2005.png
It seems to me that the pronunciation that you need to use depends on how you intend to use Greek.
If you just intend to read and write, as long as each grapheme is distinct it doesn’t matter how you pronounce them. To be able to convert the letters into some sound in your head is an important part of learning a word . ie phonemes suffice.
If you want to upload videos to youtube then you have to conform to other peoples expectations
(even when those expectations are wrong).
I don’t really get poetry but that I imagine is the one area where it is important to be able to get the exact sound of the poet would have spoken.
I checked Petit Robert too, which I think is the best French monolingual dictionary. They don’t seem to distinguish different quantities of ɛ and they explicitly call mettre and maître homonyms. Check their introduction under “Evolution du système vocalique”.
Which is one of the reasons why I honestly can not be bothered. See, if it was someone who has actually spent some time on the topic criticism would be interesting. If someone has an aeshetic reason, I actually find that interesting too. However, when you get these idiots who haven’t even seen a book on linguistics commenting with the surety a panel of experts would not dare have on any subject it annoys me. Look over some of the good videos on YT to see the comments I mean “you bro beta is vita, c is actually che” and so on.
You know, I thought of putting something up here of me reading the vowels but I can’t work out the best way to do it. Youtube seems a silly idea.
OK, what was tricking me is this: when the ɛι dipthong gave way to a pure vowel in the fifth century, it closed quite a bit. Makes perfect sense because of the iota. But I was led astray by a certain author’s often deeply annoying pedagogy. Said author will remain nameless.
Now I can set up a correspondence with French–although Paul probably won’t like it:
ɛι=/eː/=(long) close mid front as in aller except a bit longer.
η=/ɛː/=long open mid front as in tête
ɛ=/ɛ/=short open mid front as in très
So, the key thing is that ɛ has more in common with η than it does with ɛι.
What would have been really cool is to hear the early Attic distinction between ɛι and ῃ, but we’ll have to leave that for another day.
C’est la vie. The Academie Frainçaise doesn’t use the IPA alphabet at all. http://atilf.atilf.fr/dendien/scripts/generic/cherche.exe?15;s=2907835125;; The more of that alphabet you know, the more of it you want from your dictionary. But then the lexicographer has to get bogged down in these kinds of issues we have been discussing and actually may not feel qualified. A happy medium has to be found. I’ve been aware of this whole matter for a long time. As I say, I learned the basic IPA alphabet (with long marks) to read my Cassell’s French dictionary a long time ago and have often found myself disappointed with dictionaries that didn’t use as much of it as I knew. Here’s the pocket Cassell’s which I am happy to report still uses the long marks: http://www.amazon.com/Cassells-French-English-Dictionary-Thompson/dp/0020136803
ἔγραψεν pster
I just checked my Langenscheidt–sad day when you have to go to Germans to learn how to pronounce French!–and they indicate ɛ: for même but ɛ for très.
γ.φ.μ.! (γελῶν φωνῇ μεγάλῃ!) lol, laughing out loud, 
They say–I don’t know if it is true or not–that the Germans are wont to say, “You’ll like Vienna. It’s like Paris, but without the French.”
I “grew up” speaking a German tradition Erasmian.
ει was pronounced as “why” and so we didn’t distinguish between αι and ει.
But, German Erasmian, American, or British, it wouldn’t much matter. I didn’t read. I decoded.
A couple years back, I picked up both a communicative approach to learning and teaching (which is not equivalent to and “auditory” method) and also the Restored Koine pronunciation (Buth, Living Koine books). Now I don’t mix up ει and αι and I wouldn’t mix up ει and η. ει is pronounced like “we” and so is iota. αι and ε are pronounced as ̈́"pet." οι and υ are pronounced with a rounded “oo” (like German umlaut uber) So I have more things I could mix up now. And I do sometimes I spell things wrong: ερχεσθε instead of ερχεσθαι, πινᾶς instead of πεινᾶς. But, it truly doesn’t matter for me now. Just as I understood the guy on the radio today saying “These things vary very much,” so I am not bothered by homonyms. I hear and understand, I don’t hear and see a visual representation of the word in my head.
For what it’s worth.
This thread on pronunciation has been surprisingly dispassionate and tolerant of difference. I suppose there is a first time for everything.
Σαῦλος: > I “grew up” speaking a German tradition Erasmian.
ει was pronounced as “why” and so we didn’t distinguish between αι and ει.
I didn’t know this. This makes as much, or as little, sense as anything. French Erasmians tend to conflate ε and η only is closed syllables. (τῆς → τες) American and Canadian Erasmians, on the other hand, tend to conflate these only in OPEN, final syllables. (φίλτατε → φίλτατη.) This makes as much, or as little, sense as anything.
ει is pronounced like “we” and so is iota. αι and ε are pronounced as ̈́"pet."
This, and the fact that you have said ἔρρωσο to the rough breathing, partly explains why in your magnificent audios you pronounce εἶναι like ἵνα. But the other reason you do this is something you have probably never even thought about or noticed, viz. the fact that most Americans pronounce all short, unstressed vowels as “uh.” (A schwa, I think they call it.) This is the way I speak English and this is the way the ESL teacher for whom I volunteer from time to time speaks English. But the funny thing is that the students, Mexicans and Iranians and Thais and Tibetan exiles who grew up in India, do NOT pronounce the English vowels this way. They actually distinguish between the vowels BETTER than we do. (the kitchen-> thee keetchen instead of thuh kitchuhn.) We never even bother to tell the students about this and no one notices and no one gives a crap because they are focused on learning the language. All of this makes as much, or as little, sense as anything.
By all means, let’s get real.
ἀληθεύσωμεν δή.
Those students are simply wrong but they right in there wrongness. They speaking as if English spelling was a guide to pronunciation. I we English speakers were sensible we would reform our spelling. These foreign students are doing the opposite - reforming English pronunciation to conform to the spelling and a result they are going to make a better job of reading and spelling than if they pronounced English correctly.
This is pretty much what Erasmus did. (That is according to Matthew Dillon in The Erasmian Pronunciation of Ancient Greek: A New Perspective, The Classical World, Vol. 94, No. 4 (Summer, 2001). I can’t pretend to have read Erasmus’ essay itself.) He simply said that all these distinctions that exist in the written language must of once reflected real differences in sound. Hence we should try and speak that pure form of the language.
There are those who argue that this pure unsullied form of the language is a myth. I really don’t know enough to judge the strength of those arguments but really - so what. If Erasmian pronunciation is a myth it is a useful myth. To read and lean the language we need to vocalize and it helps if our vocalization reflects the distinctions found in the spoken language so much the better.
Checking out Christophe Rico’s site I see he pretty much says this. He is attempting to teach 1st cent CE Greek by getting us to speak Greek. Yet the pronunciation he teaches is not 1st century. The reason he does not is that for him speaking Greek is ultimately just a means to an end - to read the texts - so it necessary to teach a pronunciation that reflects the written language.
I put that badly as I was describing from my own point of view. At the moment I can pronounce Greek in any way I want - no one can hear me except maybe the neighbors. Should I work out how to make videos uploadable to youtube then that freedom will go and should I get to the point then I will simply find out what others are doing and attempt to conform to that.
But anyone who is speaking in form that others can hear, whether troll infested youtube or an academic seminar does at least to some extent have to take into account what others expect.
Where’s the “No, and I don’t see the reason to do so” option in the poll? 
I hadn’t thought of that, but some others here seem sympathetic to that way of thinking. That leads one to ask if there is a really reason to distinguish any vowels at all. Indeed, it seems we really could just make them all schwas! Think of how much simpler everything would be!
menen eede thee Peleeedee Echelees
eeleemenen, e mere Echeeees elge etheke
Pster:
IreneY wrote:Where’s the “No, and I don’t see the reason to do so” option in the poll? >
I hadn’t thought of that, but some others here seem sympathetic to that way of thinking. That leads one to ask if there is a really a reason to distinguish any vowels at all. Indeed, it seems we really could just make them all schwas! Think of how much simpler everything would be!menen eede thee Peleeedee Echelees
eeleemenen, e mere Echeeees elge etheke
nuh uhs the tuhm fuhr uhll guhd muhn tuh come tuh the uhd uv thuhr cuhntruhmuhn.
Oh c’mon! That’s not an argument worthy of a person who studies ancient Greek! The distance between not differentiating between epsilon iota and eta and having only one vowel is huge.