<i>, <e:>, and <e> (and <u>, <o:>, and <o>) were all enunciated closely together, but in different positions. We know this for several reasons. The first is that they all merged into one phoneme for back and front respectively. Second, in Greek transliterations short i is often represented by epsilon rather than by iota, which it would always be transliterated as if there were only a difference in quantity between long and short i. Third, the Latin grammarians describe the various phonemes as having different qualities.
Babelfish wrote:Ancient Greek also had a difference in quality afaik, for Ο and Ε only - the long variants of which were therefore marked Ω and Η. Latin didn't bother to provide us with such information
In Attic Greek, the difference between ο, ω and ε, η was one of quality only, [ɔ, ɔ:] and [ɛ, ɛ:] respectively. ου and ει originally represented both [oʊ, o:] and [eɪ, e:] respectively, but then each monophthongized into their respective long monophthongs and were later raised to [u:] and [i:] during the Hellenistic period. The other vowels exhibited only differences in quantity.
Most men are rather stupid, and most of those who are not stupid are, consequently, rather vain.
-A.E. Housman