Picture of Szabolcs Horvát
Dialects differentiating between ι, υ, η, etc.?
by Szabolcs Horvát - Wednesday, 17 June 2009, 04:01 AM

In the "official" pronunciation of Modern Greek, there are several letters/digraphs that are pronounced the same, e.g. ι, η, υ, ει, οι; ε, αι; ο, ω.

When did the pronunciation of these sounds/diphthongs merge during the evolution of Greek? Are there any dialects still spoken today in which the difference in the pronunciation of some of these is still maintained (so that the spelling difference would make sense for these dialects)?

As irregular as they are, some English spellings do make sense from this point of view. E.g. in most pronunciations, bite and night have the same ending, but there are dialects (e.g. in Scotland) which still preserve the difference today.
Picture of Christophe Grandsire-Koevoets
Re: Dialects differentiating between ι, υ, η, etc.?
by Christophe Grandsire-Koevoets - Tuesday, 30 June 2009, 09:45 AM
  The only thing I can think of is the Tsakonian dialect (which is sometimes considered a separate language), where Ancient Greek υ became /u/ (Modern Greek ου) rather than /i/ (for instance, in Tsakonian wood is /'ksulo/ <ξούλο> rather than /'ksilo/ <ξύλο>).

Otherwise, all dialects of Modern Greek featured the same mergings of sounds. This happened relatively early in the history of the Greek language. By the turn of the first millenium AD, the mergings were probably already completely done. In other words, all those letters and digraphs have been pronounced about the same way for at least a millenium.

So, your search for sense in the keeping of the various spellings can stop here. There are no dialects that still make a difference between those different letters and digraphs (Tsakonian, as I have mentioned, is often seen as a different language, as it is not completely understandable by someone who speak standard Modern Greek).
Picture of Szabolcs Horvát
Re: Dialects differentiating between ι, υ, η, etc.?
by Szabolcs Horvát - Wednesday, 1 July 2009, 05:50 AM
  Thanks for the reply, Christophe!

I wasn't trying to find any sense in the bewildering variety of spellings of the Greek /i/, I was just curious smile (Though as I am progressing with grammar, some of those spellings do start to make sense.)

The main reason why I thought that the distinction between at least some of these phonemes might have survived is that in a different audio lesson I was following it was pointed out that there is a difference between the pronunciation of the O sound in οκτώ and τώρα. However, I had a very hard time hearing the difference in those recordings, and I am still not sure that I can hear it. Unfortunately the lesson was audio only, and did not mention spelling at all (and unlike the LGO series, it didn't have many explanations ... now that I finished that lesson series, I wouldn't recommend it, or any other lessons that lacks grammar explanations, to anyone). Nevertheless, since the lesson made no reference to spelling, the distinction between those two O sounds might not be related to ο/ω at all.

Picture of Guest User
Re: Dialects differentiating between ι, υ, η, etc.?
by Guest User - Wednesday, 1 July 2009, 11:56 AM

It is not only impossible but undesirable for human beings to produce every vowel of their language with exactly the same quality and quantity in every phonetic context but these sophisticated features of speech are not things to be taught as basic phonemes to someone who is trying to master the modern Greek alphabet.

The As in English 'cackle' and 'pappa' are not identical but your average English speaker is completely unaware of this tiny difference.

I so far haven't found any reason to regard omikron and omega as letters representing different phonemes and I advise you ignore this point of your audio lesson.

Picture of Greg Brush
Re: Dialects differentiating between ι, υ, η, etc.?
by Greg Brush - Wednesday, 1 July 2009, 01:29 PM
  In ancient times (approx. 600 B.C. until perhaps as late as the 1st century A.D.) written ω represented spoken /o:/, a slightly longer (prolonged in time) /o/. However, in Modern Greek there is no difference between the sound represented by ο and ω.

Let me repeat that: in Modern Greek there is absolutely NO difference between the sound represented by ο and ω.

In other words, the lesson you refer to either
1.) was referring to the sounds these letters represented in the ancient/classical language, or
2.) was completely wrong about a distinction in sound if referring to the modern language.

Hope this helps,
Greg Brush
Picture of Christophe Grandsire-Koevoets
Re: Dialects differentiating between ι, υ, η, etc.?
by Christophe Grandsire-Koevoets - Thursday, 2 July 2009, 11:49 AM
  As Greg Brush already pointed out, in Modern Greek there is absolutely no difference between ο and ω. They refer to exactly the same phoneme /o/.

However, in practice sounds are not uttered in isolation. In the speech stream, different sounds will influence each other, and the position of a phoneme in a word will have an influence on how it's actually pronounced. In your examples, we have οκτώ, where the /o/ phoneme is unstressed and in a closed syllable (a syllable ending with a consonant) and τώρα, where the /o/ phoneme is stressed and in an open syllable (a syllable ending with a vowel). Because those environments are different, the same /o/ phoneme will end up being pronounced slightly differently (typically, a stressed vowel will be pronounced slightly longer and closer than an unstressed one, and a vowel in a closed syllable will be slighter more open than a vowel in an open syllable). But those are automatic, unconscious mouth and tongue movements that produce those slightly different sounds. They are not perceived as different by the Greek speakers, and they certainly are not influenced by the actual spelling of the sound (in other words, a stressed /o/ in an open syllable will be pronounced the same, no matter whether it's spelt ό or ώ).

Maybe it was what your audio lesson was referring to when it pointed out that there was a difference between the pronunciation of those two /o/ sounds. But even if that's the case, I'd still advise you to ignore it, because in all likelihood you are already making the distinction in your speech (as I wrote, the mouth movements that induce this are automatic. They come from your mouth getting ready for the next sound in your utterance), and even if you're not, those distinctions are not consciously perceived by Greek people anyway. So it's useless to try and reproduce them (unless you want to pass as a native Greek speaker, but I don't suppose that's your goal big grin).
Picture of Szabolcs Horvát
Re: Dialects differentiating between ι, υ, η, etc.?
by Szabolcs Horvát - Friday, 3 July 2009, 04:49 AM
  Thanks for the replies everyone!
Picture of Szabolcs Horvát
Re: Dialects differentiating between ι, υ, η, etc.?
by Szabolcs Horvát - Sunday, 5 July 2009, 04:03 PM
  I found some interesting readings on the topic here:


Unfortunately only the first chapter is available in English.

According to the page, there are dialects (not only Tsakonian, but some Koine-derived too) in which υ and οι are pronounced as /u/ rather than /i/. I also found interesting the part about geminates---another feature of the language that is marked in writing, but is not preserved in standard spoken Greek.

The site has other useful pages too, among them some dictionaries (the same ones that are accessible at komvos.edu.gr, but with nicer interfaces and English descriptions). Unfortunately most of the content is not translated into English, so it is not (yet) accessible for me.

EDIT: While reading a bit about the more interesting sounding dialects mentioned in that article, I found a paper mentioning another preserved vowel distinction. Since the paper is not accessible without a subscription, I'll copy the relevant paragraph here:

"Together with the scattered Greek-speaking communities of Cappadocia in central Asia Minor, the Pontians seem to have been cut off from the rest of the Greeks by the invasions of the Seljuk Turks into Asia Minor around the eleventh century. From there on, the spoken language of the Pontians followed for the most part a separate development from the other Greek dialects. Although Greek had certainly begun to branch off into different dialects even before the eleventh century, the differences between Pontic and the rest of the Greek started to become more acute from about this time. So of all the features of Pontic that strike mainland Greeks as ancient, most were still alive all over the Greek-speaking world in medieval times, but have subsequently died out in the mainland dialects. One very significant exception is the Pontic pronunciation of the Greek letter η as 'e'; this seems to preserve the ancient pronunciation of the letter (a long 'e'), whereas all other varieties of modern Greek pronounce this letter as an 'i'. The problem for the linguist is only used in certain words and endings, while in others the letter is pronounced 'i'. No plausible explanation has yet been offered for this."