Talk:Ancient Greek phonology

Editors who are interested in improving this article are encouraged to read this talk page discussion and the previous discussion at the Archive 1 and Archive 2.

articles that we need to check

Please use this section only for things that we need to look up, that might be relevant to this article +MATIA ☎ 20:34, 23 November 2005 (UTC)

• George Babiniotis, The question of mediae in Ancient Macedonian Greek reconsidered (this study is also at ISBN 1556191448).
• A History of Ancient Greek - From the Beginnings to Late Antiquity. Edited and translated by A.-F. Christidis, University of Thessaloniki, Greece http://www.cambridge.org/uk/catalogue/catalogue.asp?isbn=0521833078 Biography: http://www.greeklanguage.gr/christidis/pubs.htm
• N. Andriotis, Greek Language History: Four essays, reprint, 1995, pp. 168 (€ 6). ISBN 960-231-058-8.
• Ανδριώτης Ν., Ιστορική γραμματική της αρχαίας ελληνικής. Μέρος Α': Φωνητική (πανεπιστημιακές παραδόσεις), Θεσσαλονίκη 1969
• Συμεωνίδης Χ., Ιστορική γραμματική της αρχαίας Ελληνικής. Μέρος Α': Φωνητική (πανεπιστημιακές παραδόσεις), Εκδοτικός Οίκος Αφών Κυριακίδη, Θεσσαλονίκη 1989.
• E. H. Sturtevant, The Pronunciation of Greek and Latin, Philadelphia 1940.
• F. T. Gignac, A grammar of the Greek papyri of the Roman and Byzantine periods. Vol. 1: phonology. Milan 1976.
• L. Threatte, The grammar of Attic inscriptions, Vol. 1. Berlin 1980.
• Sven Tage Teodorsson:
  • "The phonemic system of the Attic dialect 400-340 BC" (Göteborg, 1974);
  • "The phonology of Ptolemaic Koine" (Göteborg, 1977); and
  • "The phonology of Attic in the Hellenistic period" (Göteborg 1978).

• Geoffrey Horrocks: "Greek: a history of the language and its speakers" (London, 1997); or
• Randall Buth: Η κοινή προφορά: "Notes on the Pronunciation System of Phonemic Koine Greek"

Summary of previous discussions

One editor, User:Thrax wants to include in this article the hypothesis that the pronunciation of Ancient Greek was very similar to that of Modern Greek, whereas all other editors have the opinion that this is a marginal view and should only be mentioned in one sentence. Andreas 03:49, 22 December 2005 (UTC)

Thrax' comment on this summary has been moved to Talk:Ancient Greek phonology/Thrax's responses to RFC comments, please see section "Block warning" below. Bishonen | talk 11:32, 26 December 2005 (UTC).

Responses to RFC

This area is for editors who have not participated in the discussion until now. Please put your responses here. Please, do not edit the individual responses, but put your comments at the end of the section. Limit your comments to a few sentences.

Try not to be confrontational. Be friendly and keep calm.

Mediate where possible - identify common ground, attempt to draw editors together rather than push them apart.

If necessary, educate users by referring to the appropriate Wikipedia policies.

Thank you. Andreas 03:49, 22 December 2005 (UTC)

Being an expert of the Ancient Greek language, I can assure the readers that the reconstructed pronunciation is indeed 100% valid, and it is found in all respectable handbooks, also in the most recent ones. Even if the pronunciation of the "dasea" as fricatives may have been anticipated in certain dialects and in certain registers in the Classical period already, nothing suggests that it was adopted universally until the Roman period. --Enkyklios 16:40, 22 December 2005 (CET)

The following discussion between Enkyklios and Thrax has been moved to Talk:Ancient Greek phonology/Thrax's responses to RFC comments, please see section "Block warning" below. Bishonen | talk 11:32, 26 December 2005 (UTC).

Anti-bloat measures

It seems Thrax can't be stopped from bloating talk pages with rambling, repetitive confrontational stuff. Maybe it's just as well for the moment, for anybody who comes in on the RfC to see immediately what's going on. But in the future I propose we should find a way of keeping talk pages reasonably clean.

Thrax, please, put your argumentation somewhere, in one place together, once. I suggest you make it a sub-page of your own talk page. Then, whenever you feel you have to remind us of some argument of yours, just link to it with one short remark here.

Else, I might adopt a policy of deleting anything you write even on the talk page if it exceeds some reasonable limit (i.e. moving it to an archive of its own, or to your talk page, and replacing it with only a link.)

I know, you're going to shout censorship now, but there's no point repeating yourself over and over again here, and you may have noticed that people have by now stopped replying to you. Lukas 08:31, 23 December 2005 (UTC)

Deleting a contributor on the talkpage is a rather extreme measure, but seeing what Thrax just did to the new sections "Summary of previous discussions" (which went from two-and-a-half lines on my screen, to 15, and is no longer a summary) and "Responses to RFC", a section specially provided for outside comment, I agree with Lucas. Thrax, nobody wants to stop you from responding to comments, even in those sections, but please do it in the way Lucas requests, or he will IMO be justified in removing your comments to a more convenient place and linking to them. Incidentally, writing in ALL CAPS is seen as shouting. Please don't greet people who are here to help (specifically at your own invitation, on the RFC page), with a shout of "WRONG!" It's just counterproductive. Please be friendly, calm, and welcoming. Bishonen | talk 10:43, 23 December 2005 (UTC)

Words unchanged in pronunciation and meaning in 35 centuries.

Putting aside the preposterous claims of Thrax, it recently occurred to me that the word μέλι(=honey) is attested in Linear B myceanean. I find it amazing that this word has probably remained totally unchanged in both pronunciation and meaning in 35 centuries! So Agamemnon would have pronounced it the same way as any speaker of Greek alive today. It would be fun to find other words like that one but I cannot think of any.Yannos 23:21, 24 December 2005 (UTC)

Block warning

Unfortunately, I see that Thrax has ignored Lukas' and my pleas above ("Anti-bloat measures") for keeping this page managable, civil, and useful, and has continued to try to shout down all opposition and add long posts to the wrong section. I'm sorry it had to come to this, but I've moved the long argument between Thrax and Enkyklios to Talk:Ancient Greek phonology/Thrax's responses to RFC comments. I don't mean to marginalize Thrax or make it difficult for him to take part in the discussion, but I don't see any other alternative to the (still less desirable) option of premature archiving of this whole page, outside responses and all. Thrax, please put any further responses you wish to make to Enkyklios or anybody else who responds to the RFC at Talk:Ancient Greek phonology/Thrax's responses to RFC comments. Feel free to post a brief note in "Responses to RFC" that you have done so, and a diff pointing to your post, but don't argue in that section any more. This is not optional, since you have been making this talkpage unusable. Stop shouting and personally attacking other contributors. You are currently violating WP:CIVIL and disrupting this page. Desist or I will block you.Bishonen | talk 11:37, 26 December 2005 (UTC).

This is deliberate censorship and the tactics of dishonest charlatans who know that their lies and falsifications cannot stand up to the evidence against them so they are now trying to prevent people form reading the evidence at all. The reconstructed pronunciation of ancient Greek is an unscientific theory based on the false notion of Germanic supremacy and everyone who has read the evidence against it and the racist slurs made by its advocates against the Greeks, and every expert in Semitic linguistics, knows it is an untenable 19th century academic fraud. --Thrax 18:45, 26 December 2005 (UTC)

The proof of this deliberate censorship and dishonest tactics is shown by Macrakis recent attempt to delete my above contribution from this page. Macrakis wants to conceal the fact that modern linguistics does not support the reconstructed pronunciation as a scientific theory but only acknowledges it as an Erasmian based convention used in teaching in English speaking countries. Modern linguists are all unanimous in the fact that there is no evidence whatsoever to support the reconstructed pronunciation and Semitic linguists support that ancient Greek was pronounced exactly like modern Greek. --Thrax 20:46, 26 December 2005 (UTC)

Crossposted to User talk:Thrax:

You have been blocked for 24 hours for personal attacks and disruption. Please note that if you continue the same actions after the block expires, I will block you for longer. Many people have recommended WP:NPOV and WP:CITE to your attention. I hope you will click on those links and spend some of the next 24 hours studying them, as well as WP:NPA and WP:CIVIL. Bishonen | talk 23:02, 26 December 2005 (UTC)

Blocking people in order to gain advantage in a content war is against the rules. --Vregamoto 01:26, 27 December 2005 (UTC)

Yes, it is. I'm not taking part in any content war, I carry no brief for either side. Kindly refer to my input on this page, and lack of input on Ancient Greek phonology (I have never edited the article, except to move it back to the consensus title when Thrax had moved it to his preferred title), before you level such a serious accusation. Thank you. Bishonen | talk 02:40, 27 December 2005 (UTC)

Vregamoto (talk · contribs · deleted contribs · nuke contribs · logs · filter log · block user · block log) has been determined, using CheckUser, to be a sockpuppet of Thrax (talk · contribs · deleted contribs · nuke contribs · logs · filter log · block user · block log). I have advised Bishonen that she may block Vregamoto indefinitely and Thrax for any time up to one month for the use of sockpuppets to continue disruptive editing practices after repeated warnings. Anybody who has a problem with this may take it up with the Arbitration Committee. Kelly Martin (talk) 15:43, 27 December 2005 (UTC)

I have blocked Thrax for three weeks for inveterate disruption, especially for using a sock to evade a block (twice) and to create a false impression of support for his position. The Vregamoto account was already indefinitely blocked for inappropriate username. Bishonen | talk 16:09, 27 December 2005 (UTC)

And you think by blocking people in violation of the rules that you can impose your POV in a content war. You must be dumber than I thought. --Yarak 16:26, 27 December 2005 (UTC)

You're on the fast track to a permanent ban from Wikipedia, Thrax. Bishonen | talk 16:42, 27 December 2005 (UTC)

You are violating Wikipedia rules in order to gain advantage in a content war by blocking people with an opposing opinion and preventing them from answering opposing claims. Your despicable tactics are not doing Wikipedias reputation for fairness and objectivity much good. --AntiCensorship 17:40, 27 December 2005 (UTC)

Are you "people" now, Thrax? Don't you ever read anything except your own posts? "Anybody who has a problem with this may take it up with the Arbitration Committee." Since you do seem to have a problem with it, you might try e-mailing User:Kelly Martin directly, via the Wiki "e-mail this user" feature. She's a member of the Arbitration Committee. Bishonen | talk 19:28, 27 December 2005 (UTC)

What credibility does she have as an honest broker. None at all because she is a person actively involved in this content dispute who has violated Wikipedia rules by blocking people who have done her no harm in order to gain advantage for one side. --WikipediaHasLostItsCredibility 20:55, 27 December 2005 (UTC)

First I'm involved in the content dispute and now Kelly Martin is? I'm beginning to understand. Nobody can block you if they're involved in the content dispute, and blocking you is in itself evidence of being involved in the content dispute? Good, all is clear. Try that one with the ArbCom. Bishonen | talk 21:11, 27 December 2005 (UTC)

I'm involved in this content dispute? Hell, I don't even understand what this article is about! Something to do with ancient Greek, but beyond that, I've no clue. Kelly Martin (talk) 22:46, 27 December 2005 (UTC)

Who are you arguing with. Someone who you've blocked so that they can't answer back. That's the only way you thought police can win arguments isn't it. What more poof is needed that your blocking violates Wikipedia rules. You're only doing it to give one side an advantage in a content war on a subject you admit you know nothing about. Say my name. --WikipediaIsAJoke 00:29, 28 December 2005 (UTC)

The evidence against the reconstructed pronunciation

These accounts need to be included in the main article.

• Chrys C. Caragounis (1995): "The error of Erasmus and un-greek pronunciations of Greek". Filologia Neotestamentaria 8 (16)
• Chrys C. Caragounis (2004): Development of Greek and the New Testament, Mohr Siebeck
• Th. Papadimitrakopoulos (1889): Βάσανος τῶν περὶ τῆς ἑλληνικῆς προφορᾶς Ἐρασμικῶν ἀποδείξεων. Athens.
• Dionysios Thrax description of the sounds of the mediae (100 BC) which implies that they cannot have been anything but fricatives.
• Semitic Linguisitic theory which implies that all proto-Sinaitic based alphabets including Phoenicians, Hebrew and Greek contained letters which were allophones for proto-Semitic fricatives and plosives.
• Semitic Linguisitic theory which states that the original Hebrew non-fricative mediae and dasea became fricatives before 280 BC and that the original Hebrew fricative mediae and dasea became breaths by 280 BC, thus proto-Semitic fricative "g" stopped being represented in the written language by 280 BC but was still recognisable as modern Greek gamma in the spoken language and represented as fricative gamma in the Septuagint translation of the Hebrew bible made in 280 BC and later became a breath in the spoken langauge. --Thrax 21:06, 26 December 2005 (UTC)

Semitic connection

Now that we have a few hours of Thrax vacations, we might try to make up our minds whether the Semitic connection is in any way worth following. I strongly suspect that Thrax has been making it up completely, but I know next to nothing about Semitic. Obviously, he has been unable to cite any scholar who actually endorses the view that the development of Semitic might provide evidence that would necessitate revisions to the received reconstruction of Greek. Is there anybody here who can verify if anything along those lines has ever been discussed in the literature? (Enkyklios, perhaps?) Of course, if it's just pure Thracian original research, there's no need bothering about it. Lukas 12:08, 27 December 2005 (UTC)

My knowledge of Semitic is also rather limited. Apparently, Thrax' Semitic arguments has two parts:

1) The Semitic letters, from which β, δ, γ were derived, could indicate the voiced fricatives [β, ð, γ]. It is true that Hebrew and Aramaic ב ד ג have fricative allophones between vowels. Yet, I am not sure that it is generally agreed that this allophony can be projected back to Phoenician or Proto-Semitic. And even if it can, it does not follow that the Greek sounds written by these letters were also fricatives because the Hebrew and Aramaic have also plosive allophones, and furthermore, a language that had voiced stops but no voiced fricatives could re-interpret foreign letters indicating voiced fricatives if that was the only solution for an adequate rendering of its own phonetic system.

2) The Septuagint trabscribe Hebrew ע (‘ayin) is transcribed with Greek γ in words like

Γομορρᾶ (עמורה) and Γάζα (עזה). It is assumed that the Hebrew sound derives partly from the Proto-Semitic uvular fricative *ġ (: Arabic غ) and partly from pharyngeal fricative *‘ (: Arabic ع). However, the Greek forms do not attest that Greek <γ> was a fricative [γ] (a language that had only [g] would probably use this phoneme for rendering a foreign [γ] as well); it would only prove that Hebrew ‘ayin was still something like [ʁ] in such words.

A language that had plosive g would never have used it to represent a sound that had virtually became a breath of near silence in the spoken language and was no longer represented in the written language. The only reason why the Hebrews would have written gomorra with a gamma is if gamma was a fricative sound which resembled a breath, otherwise they would have left it out altogether. In fact almost every consonant in the Hebrew bible is translated into Greek using the fricatives gamma, deta, beta, hi, th, and fi even though the Greek alphabet have plosive k, t and p which could have been used instead as can be seen from these examples.

Hebrew, Greek, English

Gen 10:1

Nh, nwe, Noah Sm, shm, Shem hm, xam, Ham YPT, iafe8, Japheth

Gen 10:2 TYRs, 8iras, Tiras MSk, mosox, Meshech TBL, 8obel, Tubal elisa YWN, iwuan, Javan MDY, madai, Madai MGWG, magog, Magog GMR, gamer, Gomer YPT, iafe8, Japheth

Gen 10:3 GMR, gamer, gomer ASKNZ, asxanaz, Ashkenaz RYFT, rifa8, Riphath TGRMH, 8orgama, Togarmah

Gen 10:4 DDNWS, rodioi, Dodanium KTYm, kitioi, kittim TRSYS, 8arsis, Tarshish ALYSH, elisa, Elishah YWn, iwuan, Javan

Gen 10:6 hm, xam, Ham KWS, xous, Cush M(TZ)RYm, mesraim, Mizraim PWD, foud, Phut KNan, xanaan, Canaan

Gen 10:7 KWS, xous, Cush sBA, saba, Seba hWYLH, euila, Havilah sBTH, saba8a, Sabtah RaMH, regma, Raamah sBTKA, sabaka8a, Sabtecha RaMH, regma, Raamah SBA, saba, Sheba DDn, dadan, Dedan

Gen 10:8 KWS, xous, Cush NMRD, nebrod, Nimrod

Gen 10:17 hWY, euaion, Hivite aRQY, aroukaion, Arkite sYNY, asennaion, Sinite

Gen 10:25 aBR, eber, Eber PLG, falek, Peleg YQtn, iektan, Joktan --AntiCensorship 17:50, 27 December 2005 (UTC)

K = kaf k = khaf (terminal) (TZ) = Tzade (tz) = Tzase (terminal) M = Mem m = mem (terminal) N = Nun n = Nun (terminal) S = Shin s = Samech H = He h = Chet A = Alef a = Ayin T = tav t = tet

Ostensibly, Thrax' "Semitic linguistic theory" has no implications for the interpretation of the pronunciation of Classical Greek and therefore there is no reason for including it into the main article. Enkyklios 16:31, 27 December 2005 (UTC)

The Septuagint is also too late to attest 5th century Attic, isn't it? --Macrakis 16:46, 27 December 2005 (UTC)

The Septuagint dates to 280 BC and the Hebrew experts who translated it learned to speak Greek as children in the time of Alexander 50 years earlier which dates their Greek to the end of the classical period. Since the Septuagint proves that the mediae and dasea were already fully formed as fricatives any changed that occurred must have began before the classical period. For example the changes that tool place in Germanic which changed p to f took at least 400 years. This evidence is enough to demolish the claims of the reconstructed pronunciation as un-provable. --AntiCensorship 17:55, 27 December 2005 (UTC)

AntiCensorship claims that "A language that had plosive g would never have used it to represent a sound that had virtually became a breath of near silence in the spoken language and was no longer represented in the written language". In Classical Hebrew and Aramaic, Proto-Semitic *ġ had indeed merged with Proto-Semitic *‘, which would perhaps be conceived as virtually silent by most non-Semitic speakers. If עמורה and עזה were pronounced as [ʕamo:ra], [ʕaz:a] (like in Standard Hebrew), the Greek transcriptions

Γομορρᾶ and Γάζα would be rather surprising whether one takes <γ> as the plosive [g] or the fricative [γ]. However, since ע is ignored in the Greek transcription in most cases (e.g. עמלק > Ἀμαλέκ, עלי > Ἡλί), the names in which it is transcribed with <γ> were most likely pronounced differently, namely [ʁ], at least in some varieties of Hebrew. Then, it could without any problems be rendered with [g] in a language which had no fricative [γ] just like Modern Greek [γ] is rendered with [g] in most European languages.

Admittedly, I am not an expert on Semitic, but I think that my account conforms with the communis opinio of Semitologists; a similar interpretation is given on the wikipedia pages on Gomorra and Gaza. By the way, if the letter ע could be pronounced like a fricative in Phoenician as well, one would expect the Greeks to use it for /g/ if this phoneme was pronounced as [γ] already around 800 BC (as suggested by Thrax) since it would have been more precise than ג which was a plosive [g] initially. The fact that ע was converted into the vowel Ο would be an argument against the fricative pronounciation of Greek /g/.

Your opinion is not something that Semitic linguists agree with. Ayin is not what the 72 Hebrews translated as fricative gamma in the Septuagint. They translated the fricative Gamma in the spoken language as fricative gamma in Greek. Even in Arabic fricative gamma is still used in the Arabic name for Gomorrah which is spelled Ghamara using fricative gamma. Why would the Hebrews translate RaMH (where "a" represents ayin) as "regma" (Raamah in English) unless gamma was totallty fricative. It would be completely ridiculous if plosive g were inserted into a word where it does not exist when the Hebrews could have used "aa" instead as in Abraam. Why would the ancient Greeks have used ayin for fricative gamma when the Phoenician alphabet already had g as an allophone for it. The Phoenician letter Ayin in the Greek alphabet became Omicron which is an "o" sound. In proto-Sinaitic script it even resembles the letter "o". All of the evidence from Hebrew translations into Greek proves that the mediae and dasea were fricatives and that is the opinion of all Semitic linguists. Look at the examples above. Where the Hebrews had a choice in 90% of all cases they chose Greek fricatives. The only time they used the Greek letter kappa was when translating the Hebrew letter Quof whereas when they translated Kaf they always used the Greek letter Hi (X) to denote that fact that it was the fricative allophone for Kaf that was being sounded in Hebrew. --AskMelegi 18:21, 28 December 2005 (UTC)

The Jewish translators of the Septuagint chose to render Hebrew רעמה with Greek

Ρεγμα because they thought that the ע [ʁ] could not be transliterated with nothing and the Greek γ was in their eyes the most adequate solution, being a voiced velar. If may have been a fricative in some varieties of Greek. My point is that even if it was always a stop, it was close enough to be used as the proper rendering of a velar fricative. And therefore it proves nothing about the actual pronunciation of Hellenistic Greek. Enkyklios 13:12, 31 December 2005 (UTC)

Fairy tales, fairy tales, fairy tales. The Hebrew translators DID NOT translate Ayin as gamma. They translated Semitic fricative g in the spoken language as Greek fricative Gamma. Ayin was translated as "o" since that is what ayin became in the Greek version of the Phoenician alphabet. The fact that Greek gamma was used to translate a "g" sound which in the spoken language was little more than a breath proves that gamma was a fricative in the Greek language. These Hebrews would have never translated a breath as hard g. A breath would have been translated as either η or a fricative and that fact that this fricative was gamma and that this gamma coincides with the proto-Semitic fricative gamma in other Semitic spellings of Gomorra is irrefutable as proof that ancient Greek gamma was fricative like it is today. --HansChristianAnderson 16:59, 2 January 2006 (UTC)

The Hebrew sound or sounds indicated by the letter ע were transliterated (not translated!) either with <γ> or nothing (not with <ο>!). In all likelihood the double treatment reflects two different pronunciations in Hebrew, namely [ʁ] and [ˤ] respectively. To repete myself, the Greek sound may have been a fricative in some or all contexts and in some or all registers, but even if it was always a stop, it would have been the most adequate letter for transliterating a voiced velar (or uvular) fricative. In the standard alphabet there was no independent grapheme for the "breath" (<η> was /e:/), and the Greek phoneme /h/ would have been inappropriate anyway since it was voiceless and quite different in force compared to [ʁ]. Enkyklios 13:19, 3 January 2006 (UTC)

I have no idea what Thrax, sorry, AntiCensorship wants to prove with his long list of Greek transcriptions of Hebrew names. Perhaps that Hebrew פ /p/, ת /t/, כ /k/ was often transcribed with <φ>, <θ>, <χ>? We already know that the Hebrew stops had fricative allophons in the position between vowels, and that it was customary for Greek to render foreign voiceless fricatives with the

δασέα. To the best of my knowledge, the Septuagint proves nothing about the nature of the Greek δασέα and μέσα. They could have been fricatives, but the transmitted forms are explained just as adequately if one maintains that they were in fact aspirated and voiced stops respectively. Enkyklios 13:43, 28 December 2005 (UTC)

It is universally accepted by all linguists that Hellenistic Greek was pronounced exactly like modern Greek. --AskMelegi 18:21, 28 December 2005 (UTC)

On the contrary: All linguists working with Ancient Greek agree that Hellenistic Greek was pronounced like Classical Greek in some aspects and like Modern Greek in other aspects. The vocalism was close to that of Modern Greek (even though the differences of quantity was not given up in all varieties of Greek until late in the the Roman age); the aspirated stops and the voiced stops may have become fricatives during the Hellenistic period, but we do not know when that pronunciation was adopted universally. But that is not the same as saying that Hellenistic Greek was pronounced exactly like Modern Greek. Enkyklios 13:12, 31 December 2005 (UTC)

More fairy tales. All modern linguists agree that Hellenistic Greek was pronounced almost identically to modern Greek with the dasea and mediae as fricatives, a stress accent and the diphthongs as single sounds and that if any changes took place in the pronunciation of ancient Greek they took place well before Hellenistic times. --HansChristianAnderson 17:03, 2 January 2006 (UTC)

Sven-Tage Tedorsson, who has done the most thorough investigations of the phonological development of Hellenistic Greek, concluded that the μέσα started to become fricatives during the Hellenistic age starting with /g/ in the 2nd century BC (documented by the occasional loss in writing); /b/ and /d/ followed only later (there is no confusion of <ευ, αυ> and <εβ, αβ>). The fricativization of the δασέα seems to be post-Hellenistic. According to Teodorsson, the diphthongs /ai/, /ei/, /oi/ were monophthongised to ɛ:, i:, ø: already in late Attic. ø: was raised to y: later, and the loss of distinctive vowel length seems to be a phenomenon of the Roman Age. (Summarised after my memory.) In other words, it is the nationalist dream of an ever unchanged Greek language that is a pseudo-scientific fairytale. Enkyklios 13:19, 3 January 2006 (UTC)

An analogy: In Russian, Hamburg is transliterated as Gamburg, but this does not mean that russian <Г> is a fricative. However, <Г> is pronounced [ɦ] in Ukrainian.

Are there any sources for the role of the Septuagint in reconstructing Hellenistic Greek phonology? Andreas 22:23, 28 December 2005 (UTC)

I wasn't involved in this initial, volatile battle, but recently my personal experience, in taking classical and "modern" greek has led me right into a similar place. I wasn't aware until recently of the real defensiveness and hostility that at least some contemperary hellenic speakers hearing any reconstructed pronunciation of ancient greek arouses. That experience, as I have just encountered, I postulate has to with perhaps merely a political, even religious confrontation. Politics and religion seem to always involve the way someone speaks. For natural Greek speakers, the words of the many reconstructed pronunciations are not the sounds they know. That much is certain. However when people are hot-headed and coming around with "the modern pronunciation is the correct one" etc. with no argument, it is of little value to an honest discussion of this subject. For one thing, ancient Greek has an aspect of meter which follows long and short syllables, for another thing, the greek pronunciation makes no distinction, and even people teaching greek don't know, the correct so called "breathing marks" used with words which begin with vowels. From these things it seems to hold the morphology of the writing doesn't reflect the greek pronunciation in current use. Therefore, because of this at least establishable difference in morphology I must conclude that the matter of pronunciation has taken on a political attribute which isn't helpful for a factual discussion of the ancient phonology. What also an be observed is that the christianization of hellenic culture had profound impacts on the nature of art and religion in greek society. by and large the orthodox religion seems to be tied closely to contemporary pronunciation. As stated before however arguments to authenticity in pronunciation based on merely contemporary greek identity seem to be based on analogy. It is also demonstrable that ancient greek had several dialects, the language of Homeros uses words that Platon doesn't and visa versa. Therefore there is an empirical demonstration of the language changing over time, or perhaps more accurately over time and within different dialects. It is unclear that in antiquity, being so unreachable because it has perished, leaving only artifacts like bones, if you will permit me a poetic metaphor, if although we can trace the morphological developments, how one dialect influenced another. Another example of this is the difference between shakespeare's english, american english, the language of Beowulf and the "boston accent" or the southern "drawl." It seems claims to know the "correct" pronunciation are posturing without basis. It also seems that ancient greek, and contemporary greek both use an "indo-european" grammar and are written left to right. to address the question of semitic language above, the grammar of any greek has more relation to sanskrit than to semitic languages, which use consonants as roots and change the vowel pattern to conjugate, while "indo-european" language relates roots to specific syllables and has other specific suffixes and prefixes it uses in conjugation. I think there is however a comparison that could be made which potentially complicates this issue a lot: even in vedic sanskrit a "sandhi", or rules of pronunciation, is expressed in a morphology. However this expression is the result of the writing used being a been invented for an oral language. sandhi is the alteration of actual sounds to make the language sound a certain way. Could "modern" greek in fact reflect a kind of "sandhi" that is not refleted in morphology, but was nevertheless used while speaking any form of greek? Could the written language have always reflected an arbitrary form, perhaps originating from a time before it was used, but even then pronounced differently? This is a question I've begun to ask myself. I don't know what evidence exists that could shed light on this, or if the comparison with sanskrit is useful. I think claims to know the correct ancient pronunciation are debatable, anymore than perhaps we should interpret the art of classical athens in relation to the tradition of byzantine art. They represent different ideas. It seems at one point it was desirable to see the naked male as a great symbol, and this was not needed later. the appreciation of male beauty seems something christianity suppressed, and continues to treat with scorn. this comparison is something i would like to ask those who believe the contemporary pronunciation must be the only correct one. would the orthodox church set up, say a naked apollon, or zeus, or dionysos in the place of christ?

I'd like to add however that there is another reason, perhaps not a logical one, which isn't represented, that seems to weigh in favor of the contemporary pronunciation. it wouldn't surprise me that the western european nations who gained political power wouldn't plunder a nation's language and set up it's utterance with its own rules, just like it wouldn't surprise me that it would do the same with with that nation's artistic treasures. the only thing to see about it is that there's obviously a schism in the traditions, and that schism is marked by the pronunciation. these two groups don't seem to have any room for the other.

Indeed there is third piece of evidence, one that points to something else. the word of water in ancient greek is, as everyone knows, hydor or idor. The modern word is nero. It seems perhaps that what Thrax says oversimplifies a subject which i quite complicated. There are a lot of possible explainations. One of which is that modern greek preserves archaisms not in classical greek, or that classical greek has archaic words replaced by others. There's not enough evidence on this subject, at least that i've seen, to really know.

Request for sockpuppet reports

Since Thrax's editing has been shown to include disruptive sockpuppets (see "Block warning" section above), I'm through assuming any good faith from him. I ask all editors of this page to please report any new apparent Thrax sockpuppets, here or elsewhere, on my talkpage. Thrax, you're in a hole, you need to stop digging. Bishonen | talk 17:06, 27 December 2005 (UTC)

Proven/assumed Thrax sockpuppets blocked so far: User:Vregamoto, User:Yarak, User:AntiCensorship, User:ZeroK, User:WikipediaHasLostItsCredibility, User:HereWeGoAgain, User:AskMelegi, User:JacobGrimm, User:HansChristianAnderson. Bishonen | talk 19:33, 27 December 2005 (UTC)

You are already violating Wikipedia rules by blocking using blocking in order to gain advantage for one side in a content war and that is unacceptable and randomly blocking other users that have no connection with Thrax is not going to do your credibility any favours. --WikipediaHasLostItsCredibility 20:51, 27 December 2005 (UTC)

My credibility is fine, thanks, how's yours? Did you e-mail Kelly like I suggested? Hers is good, too. Bishonen | talk 21:05, 27 December 2005 (UTC)

Note: disruptive socks can be reverted without consideration of 3RR

It is only by evading his block through the use of abusive sockpuppets that Thrax is currently able to edit Ancient Greek phonology. Any edits to the article by the sockpuppets listed above can be reverted by any editor, without concern for the 3RR, since that rule does not apply to reverting obvious vandalism. Bishonen | talk 22:05, 27 December 2005 (UTC)

To you disruptive is anyone who does not agree with your point of view on a subject that you know nothing about. Say my name. --WikipediaIsAJoke 00:30, 28 December 2005 (UTC)

Ok. Thrax. Bishonen | talk 00:45, 28 December 2005 (UTC)

alveolar or dental?

Can anyone point to a source discussing whether τ, δ, θ were alveolar (as claimed in the article page) rather than dental, as they are in modern Greek. Dental [t] and [d] (as in Italian, Greek, French) are very close to alveolar [t] and [d] (as in English) but still there is a detectable difference (Italian accent in English is often made fun of using the mafiozo phrase: Are you talking to me? (with a dental t)Yannos 01:47, 30 December 2005 (UTC)

Answering my own question, I found that Allen believes τ, δ, θ to have been dental, as in Modern Greek, not alveolar, as in English. So I am going to change the article accordingly ok?Yannos 05:28, 30 December 2005 (UTC)

Allen believes ? Who is Allen but a redactor of 19th century literature like Hans Christian Anderson and Jacob Grimm. He is not a linguistics researcher and has never published any papers of original research on ancient Greek pronunciation so why should we take his word for anything. Semitic linguists all agree that ancient Greek was pronounced exactly like modern Greek as do a sizeable proportion of Indo-European linguists. --HansChristianAnderson 17:09, 2 January 2006 (UTC)

Can I guess who you are , let me think: Thrax?? Right! Please thrax, just show us some credible and verifiable sources of your original research and we can discuss all the matter again --Philx 01:52, 3 January 2006 (UTC)

So, W. Sidney Allen "is not a linguistics researcher and has never published any papers of original research on ancient Greek pronunciation". Yet, besides the meticulous, balanced and authoritative monograph on Greek pronunciation, he has published the book Accent and Rhythm (Cambridge 1973) and the articles "Some problems of palatalization in Greek", Lingua 7 (1958), pp. 113-133, "Some remarks on the structure of the Greek vowel system", Word 15 (1959), pp. 240-251, "Prosody and Prosodies in Greek", Transactions of the Philological Society (1966), pp. 107-148, "A problem of Greek accentuation", in: Bazell, Catford, Halliday & Robins (eds.), In memory of J. R. Firth (London 1966), pp. 8-14, "The oral accentuation of Greek", Didaskalos II2 (1967), pp. 90-99, and "The Development of the Attic vowel system. Conspiracy or catastrophe?", Minos 20-22 (1987), pp. 21-32. Furthermore he has published books and articles on Latin and Sanskrit phonology.

Once again Thrax makes unfounded allegations and projects his own insufficiency on others. The fact is that both the specialists of Ancient Greek and the Indo-Europeanists all agree that Ancient Greek was not "pronunced exactly like Modern Greek" (which would be an absurd allegation), but had its own phonological and phonetic system somewhere between Proto-Indo-European and Modern Greek. Thrax continues to assert that his excentric opinion has support from a majority of linguists even though he has not been able to cite one single modern author specialised in Ancient Greek. Quousque tandem patientia nostra abutere? (P.S. it is Andersen, not Anderson.) Enkyklios 12:50, 3 January 2006 (UTC)

Caragounis' paper

I beg your pardon, but can someone point me at any linguistic resources that refute Caragounis' study? I know Caragounis is not a linguist, but his research has asked some interesting questions, which I haven't seen answered by anyone (but I haven't searched much). So could someone point me to some text that scrutinises Caragounis' paper? I know most of the linguists have not even commented on Caragounis or outright dismissed his views, but so were they initially doing for Erasmus. 195.173.80.1 16:25, 5 January 2006 (UTC)

I'm afraid I'm not aware of a published refutation in the linguistic literature either. But several fairly competent linguists among the contributors here have at least hinted at reasons why they find Caragounis' work unsatisfying, and to be honest, much of what is objectionable in it is really self-evident to the more-or-less-expert reader. Of course, that's not the stuff to write into the article, as it would be "original reasearch", but you can be assured it is a well-informed judgment. I'd be happy to provide a summary of my own thoughts, perhaps on my talk page, if you asked me to do that (and if I find the time). If you want to discuss this further, would you mind providing your username (or getting one)? Somehow it feels nicer to discuss if there is a name you can address. Lukas 16:44, 5 January 2006 (UTC)

I should have added that there have apparently been a couple of reviews of Caragounis' 2004 monograph. A list is here: [[1]]. Another one, by Moises Silva, due to have appeared in Westminster Theological Journal in autumn 2005, was apparently going to be scathingly negative, according to pre-published excerpts here: [[2]]. A response by Caragounis himself can be found here: [[3]]. All the reviews so far seem to be in the theological, not the linguistic, domain, and I haven't seen any of them and don't know to what degree they address the linguistic issues. Lukas 18:37, 5 January 2006 (UTC)

Hi and thanks. I haven't decided on a user name and one more password to remember yet (my usual username is taken), but I will as soon as I find something I can contribute to. I am mainly interested on the spelling errors Caragounis has identified, and especially the ones where EU is misspelled as EB. I would like know how others think regarding this and how could it be justified because I found it very interesting. I am sure that at some point Ancient Greek was pronounced the way that is prevelent today, but not sure what point in time they changed. Cheers! 195.173.80.1 11:14, 9 January 2006 (UTC)

Okay. Maybe I shouldn't have done this ;-) but I've actually written down my views on Caragounis now. It's got a bit longer and more vitriolic than I originally planned. For the moment, you can find it at [4]. If anybody wants to comment, please feel free to use my user talk page. Lukas 22:23, 9 January 2006 (UTC)

I disagree with this article using such a bad scholarship as Caragounis. I gave up reading his paper when the "proves" ancient ευ had the sound of ef/ev, because Latin inscriptions show ev, only paragraphs after saying Latin phonology was an useless approach to greek phonology, he just forgot to mention that v only appeared in the Latin alphabet in the XVI century, and that it is above all doubts that what is v in modern languages was a "w" sound. I don't think any serious scholar would care to write a rebuttal to such a thing.Bruno Gripp 14:41, 10 September 2006 (UTC)

pronunciation of the diphthong oi

I found two authors (Sydney Allen and Michel Lejeune) who claim that the diphthong οι underwent the following evolution in pronunciation: [oi]>[øi]>[øː]>[yː]>[y]>[i]. In classical times it would have been [øi] already. The monophtongization to [øː] would have occurred in roman times and the closing to [yː] soon thereafter (starting to be confused with upsilon). Then, the story goes, the loss of vowel quantities accross the board (second-third century AD) reduced the sound to [y], with the modern sound [i] being set no earlier than the 9th century AD! Should we include all that in the article, or is it becoming a long article already? Yannos 07:03, 4 January 2006 (UTC)

I guess we shouldn't load the article with too much detail. Of the 4 intermediate stages you mention above, one [yː] is already explicitly mentioned, one (the loss of length) is implied in the sentence preceding it. Monophthongization is the explicit subject of the whole paragraph, and that something like [øː] would be another stage on that way is pretty self-evident, isn't it? Perhaps we can find a way of adding a bit more precision to the wording, but I wouldn't recommend adding more than at most one more sentence for that purpose. Perhaps I'll give it a try later. Lukas 07:45, 4 January 2006 (UTC)

To complicate the case even more, [oi] and [øi] were perhaps in fact [ɔʏ] (like German eu) and [œʏ]. Anyway, the monophthongisation to [øː] was probably beginning already in the Classical period (Teodorsson) and should perhaps be included in the article. Enkyklios 09:23, 4 January 2006 (UTC)

The detailed phonetic changes are interesting, but less important, I think, than the structural changes. The crucial stages are those which change the phonemic inventory of the language, or which change the position of a sound within the overall system, e.g. from long to short or diphthong to monophthong. They are also the changes for which we have the best evidence. --Macrakis 15:01, 4 January 2006 (UTC)

This probably is why many Germanic languages (and a few others) have Oekologi or something similar instead of Ecology. Did /ai/ go under a shift which involved /æ/?Cameron Nedland 22:28, 7 November 2007 (UTC)

I'm dubious about the theory, but in any case, I think it's safe to say that the Germanic languages are adequately explained as ultimate consequences of the equivalences Greek oi=Latin oe, Greek ai=Latin ae. Wareh 00:57, 8 November 2007 (UTC)

Diphthongs au etc.

Should it not be [aʊ] etc as in Greek alphabet? Andreas 22:57, 12 January 2006 (UTC)

As it is well-known, the diphthong /au/ is normally pronounced [aʊ] in English and German, i.e. the second vowel of the diphthong, [ʊ], is slightly less closed and less back-tongue than the "clear" archetypical [u]. The fact that Greek /au/ ultimately becomes [av, af] speaks against a more open quality of the second vowel. On the other hand, misspellings like

ΑΟΤΟΚΡΑΤΗΣ for Αὐτοκράτης (Attica, IG II(2).10706) indicate that [aʊ] may have been a possibility in some varities of Greek. As a matter of fact, the non-standard spelling <αο> for <αυ> is rather frequent in inscriptions from the Ionic-speaking cities in both Asia Minor and the Black Sea area, which indicates that [aʊ] was in fact a typical feature of Ionic. (An alternative explanation would be that the scribes considered <ο> more appropiate for [u] than <υ>, which normally stood for [y].) I think it is better to avoid the more specific [aʊ] and stick to the more general [au] given that our sources give no unequivocal indication of the proper pronunciation. But it should be the same in the two articles. Enkyklios 08:20, 13 January 2006 (UTC)

Dasea

The french article has /d̥ʰ/ etc. for <θ> etc. No sources are given there. Comments? Andreas 23:01, 12 January 2006 (UTC)

That's slightly silly because in IPA /d̥ʰ/ is an aspirated voiced dental plosive, but the circle denotes that the voicing is removed so we get an aspirated unvoiced dental plosive, i.e. /tʰ/. So there is no disagreement with the French as to what the pronounciation was, but they have chosen an awkward way of representing it. Stefán Ingi 00:20, 13 January 2006 (UTC)

I am not sure that d̥ = t. The circle denotes devoicing, which is not the same thing as inherent voicelessness. I would normally use d̥ if I wanted to point out that it is an allophonic variant of plain d. I agree that in the context of Greek this is inapropriate.Yannos 06:56, 13 January 2006 (UTC)

In my own langauge, Danish, the old opposition between voiceless and voiced stops has been given up in favour of an opposition between aspirates [b̥ʰ, d̥ʰ, g̊ʰ] and non-unaspirates [b̥, d̥, g̊] (voice being irrelevant and absent in the normal pronunciation). The stops in question do not have the muscular tension characteristic of the fortis, but at the same time they lack the voice normally associated with the lenis. In Ancient Greek, however, where we have a system of three articulations, voiceless, aspirated and voiced stops, the notation [b̥ʰ, d̥ʰ, g̊ʰ] is probably misleading since it would be natural to articulate the three rows with the greatest possible distance. Enkyklios 07:47, 13 January 2006 (UTC)

Does anybody know any sources that back [b̥ʰ, d̥ʰ, g̊ʰ]? Andreas 21:53, 15 January 2006 (UTC)

I suspect that the reason for introducing the concept of voieceless aspirated lenis into Ancient Greek is that one gets a symmetric maximum contrast in the consonant system:

	tension	voice	aspiration
π τ κ	+	÷	÷
β δ γ	÷	+	÷
φ θ χ	÷	÷	+

By that means, one also has the opportunity of putting the later parallel development of the

δασέα and μἐσα into fricatives on a common formula, viz. [÷tension] > [+fricative]. However, the δασέα was in all likelihood fricativised hundreds of years later than the μἐσα. Furthermore, the very category of the voiceless lenis is disputed. Thus, in Paul Laver, Principles of Phonetics (Cambridge 1994), a rather comprehensive handbook of phonetics, we are told (pp. 343-4):

"Second, a question is prompted about whether devoiced and voiceless segments are phonetically identical in all respects. One possible ground for maintaining that they are not identical would be that when a segment is fully voiced, various phonetic features (over and above the phonatory difference) have different values than when the corresponding voiceless segment is produced. These differentiating factors may include, potentially, the degree of muscular effort being exerted throughout the vocal system, particularly in the muscle system of the supralaryngeal vocal tract. Such differences of overall muscular effort have been posited as the basis for a distinction that is sometimes drawn between fortis sounds with high overall muscle tension (normally voiceless) and lenis sounds with low overall muscle tension (normally voiced). The argument would then run on to say that if the only change from a lenis voiced segment in producing its devoiced counterpart were a change of phonatory state from vibrating vocal folds to an open glottis, then the lenis devoiced segment would still be differentiated from its fortis voiceless counterpart by factors of overall muscular tension.

"The fortis/lenis distinction remains to be confirmed empirically, however, and until it is, a more cautious position would be to accept the equivalence of fully devoiced and voiceless sounds, as a limit to the descriptive power of phonetic theory. Both categories should then be treated as instances of voiceless sounds."

At any rate, even if we accept [b̥ʰ, d̥ʰ, g̊ʰ] and [b̥, d̥, g̊] in Danish and Icelandic (see Talk:Danish phonology), there is no reason for assuming that the Greek voiceless aspirated plosives must have been lenis as well (unless one denies that the aspirated tenuis exist). I am quite sure that no inscription has yielded the digraphs <βh, δh, γh> (like the frequent <πh, τh, κh>), which would be expected if the

δασέα were somewhere between the μἐσα and the ψιλά. As far as I know, the only source to make this assumption is the French wikipedia page, and it hardly suffices in this context. It is, therefore, more in accordance with the principles of wikipedia and also "a more cautious position" to transcribe the δασέα with the IPA letters [pʰ, tʰ, kʰ]. Enkyklios 09:09, 16 January 2006 (UTC)

How about Hindi [b^ɦ] as opposed to [p^h]? Andreas 14:47, 20 January 2006 (UTC)

Indic bh, dh, gh are normally described as voiced stops with a breathy-voiced release. The proper IPA translation is [bʱ, dʱ, gʱ]. The Danish aspirated stops [pʰ/b̥ʰ, tˢ/d̥ˢ, kʰ/g̊ʰ], on the other hand, are never voiced. Enkyklios 11:29, 22 January 2006 (UTC)

Sorry, I've never been able to find enough time to finish the French article and to quote my sources: Michel Lejeune, Phonétique historique du Mycénien et du grec ancien, Éditions Klincksieck, mainly §56 and §61 about the fact that, for him, so-called aspirated consonants were lenis consonants, that were distinguished from fortis (unvoiced stops, for instance) ; hence φθ, χθ (lenis [de]aspirated + lenis aspirated) are not equivalent to πφ or κχ (gemminated consonants). See also §42 (). Michel Lejeune quotes epichoric inscriptions that confirm his views. If Lejeune is the only one to consider aspirated as lenis consonants opposed to fortis ones, I suppose that should be corrected. Vincent Ramos 16:39, 11 February 2006 (UTC)

Knacklaut

Germans think that spiritus lenis was a de:Knacklaut, i.e. a glottal stop, [ʔ], presumably because this sound exists in German phonology. Is this supported by credible sources? Andreas 15:06, 24 January 2006 (UTC)

Greek versification speaks against a knacklaut. The liaison of a final consonant and the elision of a final vowel suggest that Ancient Greek behaved like most languages (including Modern Greek):

μῆνιν ἄειδε θεὰ Πηληιάδεω Ἀχιλῆος

οὐλομένην ἧ μυρί’ Ἀχαιοῖς ἄλγε’ ἔθηκε

Enkyklios 17:33, 24 January 2006 (UTC)

Greek language article series

Could people interested in this article please have a look at a discussion I instigated at Talk:Greek language, regarding a proposed restructuring of the whole series of Greek-related language articles. Thanks! Fut.Perf. ☼ 07:32, 1 June 2006 (UTC)

illogical accent rule

The following is illogical and probably incorrect, as far as i remember what i learned in college:

The accent can fall only on one of the last three syllables of a word, and if the last syllable contains a long vowel, it can fall only on one of the last two syllables. (Hence the circumflex can only fall on the last two syllables.) --Espoo 11:42, 10 September 2006 (UTC)

why is this illogical? isn't accentuation, or indeed in ancient greek PITCH always artrirary, and conforming to the given rules of a language? —The preceding unsigned comment was added by 69.177.131.120 (talk • contribs).

Actually, Espoo is right - it's the "hence" that's illogical. The second statement doesn't strictly speaking follow from the first statement alone. But each of the two statements in itself is correct. Fut.Perf. ☼ 15:52, 9 October 2006 (UTC)

Nomenclature

I'd like to see someone more knowledgeable tackling with all the crazy divergence in phonetic nomenclature circling round the consonants, which will drive to tears any student dealing with the third declension in different grammars, particularly grammars in different languages. For instance, Georg Curtius puts daimon with the dentals; in a portuguese language grammar I have the third declension consonant groups divided in oclusives (or obstruents): labials (pi, beta, phi), dentals (t, theta, delta) and guturals (k, gamma, X); then liquids (rho and lambda), nasals (n and m) and sibilants (sigma). With Goodwin a whole other matter etc. Has any grammarian made something out of this mess; or is there any more authoritative classification in currency?

201.19.179.56 00:35, 18 October 2006 (UTC)

Are Ksi & Psi affricates or just biphonemic clusters?

The article seems not to mention about a possible affricate nature of Ksi & Psi as I've often read in atricles. Is there any reason for that? --Koryakov Yuri 16:06, 29 October 2006 (UTC)

You're quite welcome to bring references to any treatments that analyse them as affricates. I don't think it's a very common claim, but I seem to remember hearing about it somewhere. Sure, if it can be substantiated it should be mentioned. Fut.Perf. ☼ 16:56, 29 October 2006 (UTC)

I'm not sure that those being affricates is even possible. Because /s/ is Coronal, /p/ is Labial, and /k/ is Dorsal.Cameron Nedland 16:35, 6 September 2007 (UTC)

My understanding is that an affricate is a stop with a homorganic fricative release. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 16:37, 6 September 2007 (UTC)

η development

Also there is nothing about when and why η became /i/, at least before or after short/long vowels merge. And also considering that ω merged with omikron and not ου. --Koryakov Yuri 16:06, 29 October 2006 (UTC)

There's now a more extensive article on the post-classical developments at Koine Greek phonology. Maybe some material needs to be merged between the articles. Fut.Perf. ☼ 16:54, 29 October 2006 (UTC)

Dialects

There is one Greek dialect that preserves eta as an eta sound and logically even keeps nu at the end of neuter singular nouns, and has other traces of the way ancient greek must have been pronounced. I consider, as those above me that an unchanging greek pronunciation or a correct greek pronunciation is somekind of dialectism (rather than nationalism) and disregards an empirical approach. It seems to me the pronunciation of the greek is something that certain people identify with prestige, and due to the many varying dialects, having been in greece recently, that the idea that there is one way to say greek is kindof like someone saying there is only one chinese and it has always been pronounced only one way and speakers of other dialects are just wrong. It is empirically a fact that there is no one dialect of greek currently spoken and written, there are many. None of them is ancient Attic, Epic, old Ionic etc.—The preceding unsigned comment was added by 69.182.213.114 (talk • contribs). [pasted from previous version by  Andreas  ^(T) 21:43, 16 December 2006 (UTC)]

Yes, Pontic Greek preserves the ancient eta as a [e] sound, that should be taken into consideration by those who think eta has been pronnounced as [i] since Ancient Greek times. Actually it seems that eta was changed to [i] in late times, well after ei merged to iota in [i]. Also, there are dialects that have [u] in place of the reconstructed [y] and [oi] of Ancient Greek (that merged into [y] until maybe X century, when other dialects changed it to [i]. Old Athenian, Maniot and other dialects have this feature --DaniloVilicic (talk) 02:39, 20 March 2008 (UTC)DaniloVilicic

Unreferenced departures from Allen in vowel charts?

These charts are introduced with, "the following scheme proposed by Allen (1968) is generally accepted." But Allen (p. 60) says the sounds represented by ε and ο are "rather like the vowels of English pet and German Gott." Thus Allen held that ε = the open-mid front unrounded vowel and ο = the open-mid back rounded vowel. Certainly his examples show that he was not suggesting the values indicated in this chart, and repeated in the articles Ancient Greek and epsilon (where it's not specified Ancient Greek or Modern Greek—but in neither, according to Allen, was ε = the close-mid front unrounded vowel). Phonology is not my forte, but I think I'm reading Allen correctly. If so, either the chart needs to be changed to follow Allen, or references should be given supporting the values given (and the claim that the chart follows Allen modified). Wareh 18:36, 10 January 2007 (UTC)

Thanks for the reminder, and sorry I forgot to answer earlier. As I haven't got Allen here at home, I can only answer (as so often) with Horrocks and Babiniotis. Both of these authors don't comment on the exact (mid-open or mid-close) phonetic nature of short <ε> and <ο>. Both use /e/ and /o/ as phonemic renderings, just as we seem to be doing here. The point is, since the short vowels didn't have a four-heights distinction, the mid ones could easily have had either quality, or could have varied between both, or simply been in between (as they are in Modern Greek). It's common practice in phonology that if a language has only a three-heights distinction, the middle members are phonemically rendered as e and o irrespective of their exact phonetic nature, these being simply the more familiar and easy to use symbols. And since lengthening and contraction of /e, o/ led to <ει, ου> (i.e. /e:, o:/, not Template:Phonetic), I was under the impression that a distinctly mid-open phonetic quality would have been unlikely at that stage. – By the way, if Allen really uses only his impressionistic analogies with "Gott" and "pet", and doesn't give a distinct phonetic description, it's actually not quite easy to reconstruct what he meant. The English "pet" vowel at least has quite a large range of possible realisations, and in some forms of RP British English it's pretty much extactly "mid". Fut.Perf. ☼ 17:58, 27 January 2007 (UTC)