Talk:Koine Greek phonology

Polytonic template needed.
Many of the examples of polyphonic Greek characters are illegible in Internet Explorer. They all need the polytonic template. --rossb 11:44, 30 October 2006 (UTC)


 * This should be fixed with my latest edit. Rnabet 15:38, 4 November 2006 (UTC)

ξ,ψ
Why on earth isn't there anything about the pronunciation of ξ and ψ? --Henri de Solages 17:20, 12 October 2007 (UTC)
 * Indeed. I notice some reconstructions over at Wiktionary presenting a short-lived spirant pronunciation /xs/, /ɸs/. Anyone want to fill in if this checks out or not? -- Trɔpʏliʊm • blah 21:43, 15 June 2013 (UTC)


 * I highly doubt that ξ and ψ ever pronounced /xs/ and /fs/. That would mean that ξ and ψ were double consonant versions of the aspirated stops χ and φ, so would have gone through the same evolution  as /kʰs, pʰs/>/xs, fs/>/ks, ps/ with the Byzantine rule that there can't be two fricatives or two stops next to each other via dissimilation.  However, most standard analyses of Greek phonology I've seen argue that ξ and ψ have always pronounced /ks, ps/.  Anyways, usually I would think that an /s/ would block the aspiration process; the /kʰs, pʰs/ pronunciation just seems rather difficult.  Geoffrey Horrocks took Old Athenian spellings of ΦΣ and ΧΣ as implying the prevention of aspiration.Iotacist (talk)Iotacist  —Preceding undated comment added 02:19, 20 May 2016 (UTC)
 * /s/ is commonly redundantly [+aspirated] (cf. e.g. the earlier Greek shift *s > h, or *sP > Pʰ in Tsakonian), so it's entirely possible for ξ and ψ to have been phonetically [pʰs], [kʰs] but phonemically simply unmarked for phonation: /Ps/, /Ks/. -- Trɔpʏliʊm • blah 15:17, 21 May 2016 (UTC)
 * The question is, though, is it commonly believed that ξ and ψ went through the change of /kʰs, pʰs/>/xs, fs/>/ks, ps/? If so, adding in discussion about these should at least be considered, but only if someone can find a sufficient source presenting reasonable evidence.  If not, we have no basis to add in a section for these.  Like I said, beyond Wiktionary I haven't (yet) seen anyone argue for the pronunciations of /kʰs, pʰs/ and /xs, fs/.

Iotacist (talk) 00:13, 27 May 2016 (UTC)Iotacist
 * Actually, rereading Horrocks's short statement on the ΦΣ/ΧΣ spellings, it seems that to him the /s/ element was thought of as at least taking the place of or similar to aspiration, c.f. Horrocks (2010: 40) I take back my remark that /pʰs, kʰs/ are difficult pronunciations, which they are not.  Nevertheless, he still does transcribe these consistently as /ps, ks/, as does nearly everyone else besides Wiktionary.  Surely they had found something.  On the other hand, if ψ and ξ were considered aspirates, subsequently fricativized, some of the evidence in Byzantine times (apart from outcomes in the modern language) for the dissimilation development of /fs, xs/>/ps, ks/, /sθ, sx/>/st, sk/, /fθ, xθ,/>/ft, xt/, etc.––i.e. υσ/ψ confusions (ἔπαψε for -υσε, 813)––is discounted because writers would identify υσ with ψ (/fs/?) anyways.  If anyone knows of any serious arguments for the /kʰs, pʰs/>/xs, fs/>/ks, ps/ pronunciations, please sat so as this deserves consideration.

Iotacist (talk) 01:20, 17 August 2016 (UTC)Iotacist

'Plosive and former plosive consonants'
Why are φ and β listed as bilabial consonants, /ɸ/ & /β/? The text doesn't seem to suggest that there has been a transition from /pʰ/ and /b/ to /f/ and /v/ through /ɸ/ and /β/.
 * I've no competence for greek but, if this can help a bit, in present Mongolian, the pronunciation of в seems to be switching from [β] to [v], specially in the capital city Ulaanbaatar. The switch has already affected в's followed by a consonnant, but never the final в's. The в's followed by a vowel seem to be pronounced [w]. в is traditionnaly described as bilabial, which seems to be the traditional pronounciation in all cases. --Henri de Solages (talk) 11:23, 10 May 2009 (UTC)


 * The intermediate bilabial fricative stages /ɸ/ and /β/ are considered hypothetical, but are still very probable. I have read another possible transitional phase for φ, and for that matter the other aspirates θ and χ as well, in Horrocks.  He points at the possibility of a short-lived affricate pronunciation of /pf, tθ, kx/, before these lost their plosive element.  They would be especially prone to simplify to /f, θ, x/ following /s/ or /f/ (or /φ/: the second element of the αυ/ευ diphthongs, please see my question on these at the bottom), or a cluster with another aspirated plosive i.e. φθ, where the pronunciations of /spf, stθ, skx/ or /fpf, ftθ, fkx/ would be difficult.  Anyways, we did note on the page that they ultimately resulted in /f/ and /v/.  Iotacist (talk) 18:45, 26 June 2016 (UTC)Iotacist

ει
The evolution of the pronunciation of ει before a vowel and before a consonant is explained, but what about final ει? --Henri de Solages (talk) 01:41, 3 March 2013 (UTC)

As a less learned visitor to this page I was confused by the time designations "late Roman period" and "early Byzantine period." Perhaps the first time these designations are used they should be explained by providing the century in parentheses. — Preceding unsigned comment added by 71.70.247.150 (talk) 10:48, 22 April 2013 (UTC)

diphthongs
It can only be on oversimplification to present a language spoken by people from Italy to India and from the Crimea to Egypt, and over a period of several centuries, often by non-native speakers, as having a defined pronunciation. Besides, the fricatization of Y in diphthongs is evident not from 1st C AD Egyptian records but already in archaic Greek inscriptions, where two or three different letters are being used in place of Y to indicate different pronunciations, at least one of which is the digamma, considered a fricative.Skamnelis (talk) 11:47, 16 April 2014 (UTC)
 * I don't think I follow this critique. The article, after all, does not present "a" defined pronunciation, but that of several specific locales at specific times.
 * The original value of bigamma, FWIW, was the semivowel *w, not a fricative. -- Trɔpʏliʊm • blah 00:05, 17 April 2014 (UTC)
 * Whether as a semivowel or fricative, the differentiation of the pronunciation of υ in diphthongs is evidenced well before the 1st C AD but I can see how that statement may be strictly speaking correct. Nonetheless, it seems odd to start the article on Koine phonology by discussing the ypsilon in diphthongs. The article should be starting with something like, Koine is the name given to the general style of the Greek language used in the Hellenistic kingdoms and during the Roman period ending perhaps with the foundation of Constantinople. It is sometimes called Alexandrian Greek, etc, perhaps with disambiguation and relevant wiki links. It was a significant departure from classical Attic and is considered to have been accompanied with changes in phonology, the use of accents, etc. The overview might better contain proposals about what might have led to the change of phonology, the degree phonology was preserved and the extent of variations in phonology. Specific examples, from Egyptian papyri, etc, in my opinion should belong to the later sections.


 * I do not also know how much can we gain by comparing Greek with Latin transliterations, e.g. OVVM is cognate with ΟΒΕΟΝ but BOS with ΒΟΥΣ. Overall, the style is more "that is how such and such was pronounced", rather than "that is how a certain scholar or school believes, such and such was pronounced in such and such place and time". For example, it is stated that zeta was pronounced as [dz] but just in the next sentence as z in classical Attic, but exactly when, in what type of material and what do scholars make out of this - an ambivalence or vulgar vs learned, or written vs pronounced, or marking a transition, how widespread outside Athens, etc. A single person probably wrote most of this article and I would rather not interfere, but I would have preferred a slightly more qualified style. Skamnelis (talk) 13:47, 22 April 2014 (UTC)

Since we are talking about Y diphthongs here, I repeat again does anyone have any ideas on the timing of their fricativization, or more precisely, if they had become fricative by the 1st century AD or if they remained semi-vocalic? Again, the [-ɸʷ, -βʷ] pronunciation which Horrocks notes as a transitional phase might be a good compromise. Please see my notes below.2602:306:C439:3150:389C:BEA:91F3:13FA (talk) 01:35, 19 July 2016 (UTC)Iotacist

Geminates
Do we think koine Greek used a geminate pronunciation for consonants written double? Q·L·1968 ☿ 20:59, 5 June 2015 (UTC)
 * Very good question. From my reading of W. Sidney Allen's Vox Graeca, I don't know the answer. Clearly geminate pronunciation was lost sometime between Classical and Modern Greek, but I haven't read anything discussing when it happened. There might be spelling clues like the clues to the loss of vowel length. — Eru·tuon 04:41, 6 June 2015 (UTC)

Yes, Gignac does lists frequent examples of geminate simplification in the papyri. I have made note of this in the consonant discussion.

Iotacist (talk) 20:57, 24 June 2016 (UTC)Iotacist

[oi] > [y:]
This development can hardly be direct, can it? I'd expect it to go via [ø:] or [øy] or [ui]. Do we know anything about that? — Preceding unsigned comment added by 93.206.132.58 (talk) 02:46, December 11, 2015‎


 * Yes. It probably did go through the intermediate phases of [øi] > [ø], c.f. Horrocks (2010: 162).  Already added this in.


 * Iotacist (talk) 21:21, 16 June 2016 (UTC)Iotacist


 * I personally suspect it was already monophthongized to a long vowel in Classical Attic, parallel to the monophthongization of ει, ου, but that doesn't seem to be generally accepted. Allen says it may have been  in Thucydides's time, but he doesn't say it with much conviction, unfortunately. — Eru·tuon 23:57, 16 June 2016 (UTC)


 * It's more likely that the monophthongisation of οι ran in parallel with that of αι, so it would have started with the values and  respectively in the 2nd century BC, well after the classical period (although intermediate values  and  or  for the classical period are not altogether inconceivable, I suppose). The monophthongisation of ει, ου was likely quite early, as early Homeric manuscripts already seem to have had it; as Iotacist says, basically. --Florian Blaschke (talk) 19:41, 16 January 2017 (UTC)


 * I agree that such was probably the case with the Attic vernacular in the 4th century BC. Teordorsson (1978) suggests that the Boeotian dialect wasn't so advanced after all, and that a highly progressive dialect was already spoken in Attica: one which completed monophthongization and dispensed with vowel length distinction.  Some of his other claims, on the other hand, seem too advanced though, like that η and υ had already merged with ι.  But an  value for oι is very plausible in Hellenistic times.  However, the diphthongs   were the first to monophthongize.  If I can recall, this was quite early in the 8th-7th centuries BC; ει, ου then acquired the values for which they are best known as : perhaps an overestimation.

By the way, does anyone have any suggestions on the status of αυ/ευ in the 1st century AD (i.e. New Testament times)? I've been working on tidying up the previously un-cited biblical Koine chart on the Koine Greek page. It originally had these diphthongs pronouncing as the intermediate value of a semi-vowel. Horrocks, on the other hand, consistently transcribes these as for New Testament Greek. Indeed, there are a several instances (counted six listed by Gignac) of confusions with αβ, εβ from the 2nd century BC to the 1st century AD (before the 4th century AD, when it is generally agreed that their fricativization was complete.) Now, Gignac does state that α(υ)oυ, ε(υ)oυ are more common before the 4th century, so the older values of  or  were probably more common. But in any case, is it still too premature to assume that this sound change was complete for at least some speakers? The labialized fricative could be seen as a compromise.

Also, one editor asked several years ago about a supposed uvular value for ρ. He or she claimed to have taken this from a website, but no source was found and they switched it back to alveolar trill. I know nothing about Hebrew/Aramaic, but was wondering if this could be a Semitism. If so, which other phonemes could've been affected? Currently, I've edited the NT Greek chart to match a more standard Koine reconstruction by Teordorsson, but it would be nice to have something on a specifically Judaean dialect, if we can find anything on this (which unfortunately, I haven't...)

Iotacist (talk) 21:31, 24 June 2016 (UTC)Iotacist
 * what do you mean by ? Does the denote an unusual lip setting or what? Mr KEBAB (talk) 21:12, 21 July 2016 (UTC)

The phonetic symbol denotes simultaneous labialization/lip-rounding. Sorry I wasn't clear about this above, but Horrocks proposes several intermediate stages in the shift of αυ/ευ to fricative pronunciations. The first stage was the second element in closing to a semi-vowel, probably along with the monophthongization of the other diphthongs. Increasingly narrower closure led to the bilabial fricative pronunciation, but still with lip-rounding, of. Once labialization was lost, this became simply. It was only a matter of time before this shifted to the dental fricative phase of, as in Modern Greek. I was suggesting that since Horrocks's New Testament transcription is found at the bottom of the Koine Greek page with the more phonetically 'advanced' pronunciation, inconsistent with the more conservative reconstructed biblical Greek table above (which had ), and since it is assumed that this sound change was complete for some speakers (probably the lower classes first) as a few spelling errors with αβ, εβ appear in the Greek papyri in the 1st century, that the  pronunciation could be a reasonable compromise. Iotacist (talk) 17:34, 22 July 2016 (UTC)Iotacist
 * Impressive, thanks! Mr KEBAB (talk) 22:27, 22 July 2016 (UTC)


 * Although the pronunciation for 'ayin may have existed in Biblical Hebrew, both it and Aramaic also have an alveolar tap or trill, so a replacement with a uvular fricative does not make sense as a Semitism. The pronunciation you seem to have in mind – of resh as  – is a peculiarity of Standard Modern Hebrew and a result of Yiddish influence, where it ultimately arrived via German. --Florian Blaschke (talk) 01:51, 23 June 2019 (UTC)

3.5.6 Sample phonetic transcription: missing word?
The beggining of the Greek sample reads: "τῇ κυρία ἀδ[ελ]φῇ Μανατίνῃ Πρώβ[ο]ς ἀδελφὼ χαίριν." It’s transcribed: "[ti cyˈria mu aðelˈfi manaˈtini ˈprovos aðelˈfo(s) ˈçerin." The [mu] part is missing in the original text (κυρία μου), or it was erroneously added in the transcription. Vincent Ramos (talk) 18:14, 19 June 2018 (UTC)


 * The μου part is in the original, and it has been added in the transcription; thanks for noting this. — Preceding unsigned comment added by 138.202.129.134 (talk) 14:57, March 26, 2019 (UTC)

Koine-era transcriptions and vowel distinctions
Here's something I have observed:

In koine-era transcriptions of greek, αι\ε and οι\υ are always merged, but ω\ο generally are not. The αι\ε merger shows that vowel length was already lost, so it seems that ω\ο had some time of qualitative difference maybe?

Has any litterature explored this question?

Βώρδαρτ

>>> The merger of ω and ο is extremely frequent in Koine documents starting from the mid 2nd century BCE in Egypt; they represented one single sound, mid rounded back vowel, /o/, with no distinction of quantity or quality. The preservation of ω in foreign transcriptions was probably just a formality. As for the Coptic case, I just consulted the Wiki table on Coptic alphabet and it shows that ω and ο are used to represent the long and short /o:/ vs. /o/ in most dialects; I'd assume that this was because Greek transcription of Egyptian started in the 3rd cent. BCE, when Greek would've still maintained vowel length distinction.

--Iotacist — Preceding unsigned comment added by 2600:1700:C2A0:9C50:E58F:C8E3:71A3:59E6 (talk) 19:51, 19 August 2020 (UTC)

Controversies
This paragraph is very interesting as an academic analysis, but it lacks quotations and isn't very suitable for an encyclopaedia in this form. Opzwartbeek (talk) 16:05, 2 July 2022 (UTC)