Modern and Koine Pronunciation

Modern Pronunciation

The main barrier to replacing Erasmian with Modern pronunciation in New Testament studies is that through a centuries-long process of iotacization, seven vowels and digraphs, which were pronounced distinctly in Classical Greek, are all pronounced exactly as iota in Modern Greek: υι = οι = ει = υ = η/ῃ = ι. Consequently, the New Testament-era words ὑμῖν (meaning, you [plural]) and ἡμῖν (meaning, we) are, for example, indistinguishable in Modern pronunciation (both are pronounced approximately \ee-'meen\).

Carlson (B-Greek, 1996): hHMIN and hUMIN at the time they were in common use were pronounced differently. However, they came to be pronounced exactly the same when the copyists were active in transmitting the text of the New Testament.⁴

Buth (B-Greek, 2000): The Greek language itself couldn't maintain the ‘classical’/‘Koine’ language with the reduced vowel system. The most prominent example is HMEIS ‘we’ UMEIS ‘you’ that became intolerable as [imis] so new forms [emis] and [esis] developed.⁵

Modern pronunciation is presented in this guide to qualify the distance between Erasmian and authentic Greek; this guide, though, is not of itself sufficient for learning Modern pronunciation. Learning Modern pronunciation would require audio aids and practice speaking, preferably with a native-Greek tutor to guide correct pronunciation.

4. Stephen C. Carlson, Re: hHMIN, hUMIN : were both pronounced the *same* ?, 09 December 1996, B-Greek Mailing List Archives, http://www.ibiblio.org/bgreek/archives/96-12/0099.html (accessed 03 February 2007).

5. Randall Buth, Modern Greek, a summary and a new direction, 17 December 2000, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2000-December/014653.html (accessed 03 February 2007).

Reconstructed Koine

An alternative to Modern and Erasmian pronunciation is Living Koine Greek advocated by Randall Buth, who has reconstructed biblical-era pronunciation from documents of that period. Living Koine is based on Modern Greek, making it authentic, but with the vowels and diphthongs un-iotacized according to the evidence in the historical documents.

Buth (Notes on Koine): We have thousands of documents from the time period whose misspellings show us which sounds were exactly the same for Koine Greek speakers and which sounds were distinct and phonemic in their ears.... Some misspellings reflect dialect differences and some misspellings are just haphazard mistakes. However, when hundreds and thousands of examples of the same kind of misspellings are found all over the Greek world, then we have found items that sound alike to Koine Greek speakers.⁶

Koine, though, does not differentiate every Greek vowel. Whereas ὑμῖν and ἡμῖν were distinguishable in the biblical era, but made indistinguishable in Modern pronunciation, conversely Koine had indistinguishable words which are artificially disambiguated in Erasmian. In the Erasmian system, only the vowel- and digraph-pairs η/ει, υ/ου, and ευ/ηυ are pronounced alike; the rest are distinct. But in historical biblical-era Koine, ο/ω, ε/αι, ι/ει, and υ/οι were merged, with η somewhere between ε/αι and ι/ει. Words such as ἔχωμεν (meaning, we should have) and ἔχομεν (meaning, we do have), which are pronounced distinctly in Erasmian, were pronounced and sounded the same in the biblical era.

Buth (B-Greek, 2000): If Luke or Paul would not have HEARD the difference between ECWMEN and ECOMEN we can feel their ambiguity and their less than felicitous wording.⁷

Buth (B-Greek, 2000): Such a system means that a person would relate to a written text with roughly the same kind of phonemic, homonymic, alliterative and assonantial grid as NT writers.⁸

6. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), The Phonemic Principle, 176.

7. Randall Buth, Modern Greek, ... a new direction-Peterson, 18 December 2000, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2000-December/014673.html (accessed 03 February 2007).

8. Randall Buth, Modern Greek, a summary and a new direction, 17 December 2000, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2000-December/014653.html (accessed 03 February 2007).

9. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), The Satus of η, 178.

Reconstructed Koine Pronunciation

Living Koine recreates Koine Greek as a Modern dialect with disambiguated vowels (but within the constraints of the historical evidence) by retrofitting the Modern pronunciation with the Koine Greek vowel equivalencies.

Buth (Notes on Koine): Phonemic Koine is close enough to Modern Greek so that Greek speakers accept it as ‘something Greek’ and ‘non-offensive’, even if not the same dialect.³

Buth (Notes on Koine): What the ... equivalencies mean is that within any particular dialect, the ω-μέγα, however it is pronounced, will be pronounced like ο-μικρόν in that dialect.... [That is,] what a traveller would hear in the majority dialects all over the Mediterranean, from Rome to Judea, from the Aegean to Egypt.²

Hobbs (B-Greek, 1996): IN EVERY PERIOD, THERE WERE MANY PRONUNCIATIONS OF GREEK; at no time was Greek pronounced the same way everywhere.¹

Μετὰ μικρὸν δὲ προσελθόντες οἱ ἑστῶτες εἶπον τῷ Πέτρῳ· ἀληθῶς καὶ σὺ ἐξ αὐτῶν εἶ, καὶ γὰρ ἡ λαλιά σου δῆλόν σε ποιεῖ..
[After a little while, those standing there came up to Peter and said, You really are one of them too – even your accent gives you away! (Matthew 26:73 NET; in Mark and Luke: he is/you are a Galilean.)]

Four vowels and diphthongs are affected: υ, οι, υι, and maybe η (un-iotacizing η is more problematic for Modern usage than un-iotacizing υ, so minimally υ is differentiated from η [ὑμῖν vs. ἡμῖν], if not also η from ι [τῆς vs. τίς]).

Buth (Notes on Koine): Modern Greek speakers accept υ and οι when pronounced [ü][y] because their ears interpret the sound as [i]. However, η as [e] [eɪ] is rejected because they hear it as [ε][e] though they would expect to hear [i]. ⁴ (See the IPA Sound Chart for the proximity of [y] to [i].)

1. Edward Hobbs, Greek pronunciation, 02 August 1996, B-Greek Mailing List Archives, http://www.ibiblio.org/bgreek/archives/96-08/0513.html (accessed 15 February 2007).

2. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 1, 176.

3. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Why Use A Phonemic Koine?, 182.

⁴⁰ Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 7, 178.

Iotacization

	Erasmian (U.S.)			Attic		Koine			Modern
	α	[ɑ], [ə]		α	[a]		α		α
	ε	[ɛ]		ε	[e]		ε, αι		ε, αι
	αι	[aɪ]		αι	[aɪ]
	ι	[i], [ɪ]		ι	[i]		ι, ει		ι, ει, η, υ, οι, υι
	ει, η	[eɪ]		ει	[eɪ?]
				η	[?]		η
				υ	[y]		υ, οι
	οι	[ɔɪ]		οι	[oɪ]
	υι	[wi], [wɪ]		υι	[yi?]		υι
	ου, υ	[u]		ου	[u]		ου		ου
	ω	[oʊ]		ο	[o]		ο, ω		ο, ω
	ο	[ɔ]		ω	[?]

note: Erasmian equivalencies result from an intention to differentiate the vowels and dipthongs in accordance with Classical Greek, subverted by the limits (in the case of anglo-Erasmian) of the English vowel system: υ as [u], for example, (the same as ου) is simply more natural for a native English-speaker than non-English [y]. (Speculation: In Greek pronunciation, since the dipthongs ει and οι eventually dropped their leading vowels becoming just [i], perhaps this indicates these dipthongs in Classical Greek were on-glides, rather than off-glides as in Erasmian? Likewise, the convergence of αι with ε and the resistance of υι to iotacization until υ itself iotacized indicate that these were off-glides?)

Koine Consonants

Living Koine with full modern constants and un-iotacized υ, οι, υι, and η is effectively a late-Koine pronunciation (second to third century CE).

Living Koine with full modern constants, un-iotacized υ, οι, and υι, and modern η is effectively an early Byzantine pronunciation (second to ninth or tenth century CE).

Certain consonants, though, may also be rolled back for even earlier Koine pronunciations.

Between the second century BCE and the fourth century CE, the consonants χ, φ, and θ where variously in transition from their classical pronunciations to their modern pronunciations. For a mid-Koine first century CE pronunciation, therefore, revert χ, φ, and θ back to their plosive pronunciations: [kʰ], [pʰ] and [tʰ].

Between third century BCE and fourth century CE, the consonants β, γ, and δ where variously in transition from their classical pronunciation to their modern pronciations. For a mid-Koine first century BCE pronunciation, therefore, also revert β, γ, and δ back to their plosive pronunciations: [b], [g] and [d].

⁰⁰ [β]: The IPA symbol [β] represents a sound very close to [v] in very: fricative like [v] but bilabial like [b] in boy. See the IPA Sound Chart for the path from [b] to [β] to [v]. (See similar note below regarding the path of φ from [pʰ] to [ɸ] to [f].)

⁰⁰ [ɸ]: The IPA symbol [ɸ] represents a sound very close to [f] in fee: fricative like [f] but bilabial like [p]. See the IPA Sound Chart for the path from [pʰ] to [ɸ] to [f]. (See similar note above regarding the path of β from [b] to [β] to [v].)

⁰⁰ [y]: The IPA symbol [y] represents the sound for Fr. u and Ger. ü.

⁰⁰ αυ, ευ: Buth (B-Greek, 2000): Currently, I use ‘v’ and ‘f’ for AU, EU and HU, though I am drifting toward using the ‘Spanish’ bilabial fricatives [β and ɸ?, or just β?] as more authentic. ¹ Buth (B-Greek, 2002): ‘ev’ and ‘av’ are also close to the Roman Koine bilabial fricatives and are the sounds I use in the reconstructed Koine pronunciation. ²

⁰⁰ [eɪ]: Buth: Living Koine Greek for Everyone has included η as a separate vowel sound.³

⁰⁰ φ: Depending on where ... Luke was on this continuum, [if he was from the north] he may have sounded like lisping when speaking f, th, ch, in Jerusalem, or [if he was from the south] he may have sounded ‘sharp’ when speaking around the Aegean ⁶.

¹⁰ Randall Buth, Modern Greek, ... a new direction-Peterson, 18 December 2000, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2000-December/014673.html (accessed 14 February 2007).

²⁰ Randall Buth, Pronunciation Help, clarification, 07 March 2002, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2002-March/020079.html (accessed 14 February 2007).

³⁰ Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), The Status of η, 178.

⁶⁰ Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 180.

Greek Alphabet

Letter	Modern Name		IPA	Modern Pronunciation		Living” “Koine
Α α	Άλφα	['al-fa]	[a]	Between English [æ] \a\ in apple and English [ɑ] \ah\ in father	Compare English \a\ and English \ah\ to Greek α
Β β	Βήτα	['vi-ta]	[v]	\v\ as in very (not \b\ as in boy)	βάλλω ~\'va-loh\ (to throw)	Maybe [β]
Γ γ	Γά(μ)μα	['ɣa-ma]	[ɣ]	\gh\ (except when γγ, γκ, γξ, γχ, γ[e], or γ[i])	λέγω ['le-ɣo] ~\'leh-ghoh\ (to speak, say)
			[ɡ]	\g\ as in good when following γ	σπόγγος ~\'spohng-gohs\ (sponge)
			[ŋ]	\ng\ before γ, κ, ξ, or χ
			[ʝ]	Palatalizes ~\y\ when γ[e] or γ[i]	(See Palatalization below)
			[ɟ]	Palatalizes ~\gy\ when γγ[e] or γγ[i]
Δ δ	Δέλτα	['ðel-ta]	[ð]	\th\ as in this (not \d\ as in deal)	δώδεκα ~\'thoh-theh-ka\ (twelve)	Maybe [d]
Ε ε	Έψιλον	['e-psi-lon]	[e]	Approximately English [ɛ] \eh\	Compare English \eh\ to Greek ε
Ζ ζ	Ζήτα	['zi-ta]	[z]	\z\ as in zone	(But see also Zeta and Sigma as ~\zh\ and ~\sh\ below)
Η η	Ήτα	['i-ta]	[i]	Same as ι	ἡμεῖς ~\ee-'mees\ (we)	[eɪ]
Θ θ	Θήτα	['θi-ta]	[θ]	\th\ as in thin (not \th\ as in this)		Maybe [tʰ]
Ι ι	(Γ)ιώτα	['ʝo-ta] ~\'yoh-ta\ (palatalized)	[i]	Like English \ee\ but higher pitched	Compare English [i] to Greek [i]
			[ʝ]	Palatalizes ~\y\ when unstressed and between a voiced consonant (other than μ) and a vowel other than [i]	(See Palatalization below)
			[ɲ]	Palatalizes ~\ny\ when unstressed and between μ and a vowel other than [i]
			[ç]	Palatalizes ~\hy\ when unstressed and between a voiceless consonant and a vowel other than [i]
Κ κ	Κά(π)πα	['ka-pa]	[k]	~\k\ as in skip (except when γκ, κ[e] or κ[i])	Compare English \k\ to Greek κ *
			[ɡ]	\g\ as in good (not \k\) when following γ	ἔγκλημα ~\'ehng-glee-ma\ (accusation)
			[c]	Palatalizes ~\ky\ when κ[e] or κ[i]	(See Palatalization below)
			[ɟ]	Palatalizes ~\gy\ when γκ[e] or γκ[i]
Λ λ	Λάμ(β)δα	['lam-ða]	[l]	~\l\ as in land (except when λ[i])	(See Dentalization of Lambda, Nu and Tau below)
			[ʎ]	Palatalizes ~\ly\ when λ[i]	(See Palatalization below)
Μ μ	Μι	['mi]	[m]	\m\ as in move
Ν ν	Νι	[ɲi] ~\nyee\ (palatalized)	[n]	~\n\ as in new (except when ν[i])	(See Dentalization of Lambda, Nu and Tau below)
			[ɲ]	Palatalizes ~\ny\ when ν[i]	(See Palatalization below)
Ξ ξ	Ξι	[ksi]	[ks]	\ks\ or \x\ as in fix
Ο ο	Όμικρον	['o-mi-cron]	[o]	Nearly AmE [oʊ] \oh\ (not diphthongal)	Compare AmE \oh\ to Greek ο and ω
Π π	Πι	[pi]	[p]	\p\ as in spike
			[b]	\b\ as in boy when following μ	ἀμπελών ~\am-beh-'lohn\ (vineyard)
Ρ ρ	Ρο	[ro] ~\roh\ (not aspirated)	[r]	Trilled (or rolling) \r\ (Scottish r, not English [ɹ] \r\)
Σ σ,ς	Σίγμα	['siɣ-ma]	[s]	\s\ as in single before vowels and voiceless consonants	(But see also Zeta and Sigma as ~\zh\ and ~\sh\ below)
			[z]	\z\ as in zone before voiced consonants (optionally \s\ or \z\ before λ)†	κόσμος ~\kohz-mohs\ (world)
Τ τ	Ταυ	[taf]	[t]	~\t\ as in stop	(See Dentalization of Lambda, Nu and Tau below)
			[d]	~\d\ as in deal when following ν	τεσσαράκοντα ~\teh-sa-'ra-kon-da\ (forty)
Υ υ	Ύψιλον	['i-psi-lon]	[i]	Same as ι (except in αυ, ευ, and ηυ)	ὑμεῖς ~\ee-'mees\ (you [plural])	[y]
Φ φ	Φι	[fi]	[f]	\f\ or \ph\ as in fee or phone		Maybe [ɸ] or [pʰ]
Χ χ	Χι	[çi] ~\hyee\ (palatalized)	[x]	\h\ as in Scottish loch (except when χ[e] or χ[i])	(See Pronunciation of Chi above)	Maybe [kʰ]
			[ç]	Palatalizes ~\hy\ when χ[e] or χ[i]	(See Palatalization below)
Ψ ψ	Ψι	[psi]	[ps]	\ps\ as in lips
Ω ω	Ωμέγα	[o-'me-ɣa]	[o]	Same as ο

Sources: Primary sources for Modern pronunciation are About the Greek Language by Harry Foundalis; Greek Language and Alphabet by Katerina Sarri; and the UCL Department of Phonetics and Linguistics' SAMPA alphabets for Greek and American English. Source for biblical-era Koine pronunciation is Notes on the Pronunciation System of Phonemic Koine Greek by Randall Buth.

*English \k\ and Greek κ have slightly different places of articulation such that Greek κ palatalizes naturally before [e] and [i].

†Foundalis: The letter λ, although voiced, does not always effect this transformation on σ.¹

1. Harry Foundalis, About the Greek Language, http://www.cogsci.indiana.edu/farg/harry/lan/grphdetl.htm#p_sz (accessed 08 February 2007), [s] →[z] in front of voiced consonants.

Modern Digraphs

Notes on Modern Alphabet and Pronunciation

a. γά(μ)μα: Spelled both γάμμα and γάμα, reflecting the pronunciation of μμ as a single letter.

b. (γ)ιώτα: Spelled both ιώτα and γιώτα (γ clarifies the pronunciation of ιώτα as ~\'yoh-ta\: γ palatalizes to [ʝ] before [i], as does ι itself between a voiced consonant, such as γ, and a non-[i] vowel, such as ω).

c. κά(π)πα: Spelled both κάππα and κάπα, reflecting the pronunciation of ππ as a single letter.

d. λάμ(β)δα: Spelled both λάμβδα and λάμδα (reflecting that the β is silent when δ is pronounced th in this).

e. μι: Modern spelling of μῦ reflects the iotacization of υ.

f. νι: Modern spelling of νῦ reflects the iotacization of υ.

g. ρο: Modern spelling of ῥῶ reflects the pronunciation of ω as ο (and also the loss of rough breathing [ ῾ ]).

h. ευ+φ, ευ+β and ηυ+φ: In Modern pronunciation, ευφ, ευβ and ηυφ effectively form double consonants: εφφ, εββ and ηφφ, pronounced simply as εφ, εβ and ηφ (Foundalis). (Ευφ occurs in six words in the GNT, including Εὐφράτης [Euphrates] and ἐυφορέω [the root of euphoria]; ευβ occurs only in Εὔβουλος [Eubulus, a proper name in 2 Timothy 4:21]; ηυφ also occurs only once, in ηὐφράνθη [an inflection of εὐφραίνω, to gladden, in Acts 2:26]. Αυφ, αυβ or ηυβ do not occur in the GNT.)

i. ηυ: Nikolaou (B-Greek, 2002): In ancient Greek, HU only occurs as the temporal augment of AU and EU, in the past tenses of verbs beginning with the diphthongs above. In modern Greek, however, there are no temporal augments at all (e.g. the simple past of AKOUW nowadays is AKOUSA -not HKOUSA).... However, ... ‘KAQAREUOUSA’ (an archaic form of Greek being in official use until 1976) had retained a lot of ancient grammatical characteristics- the temporal augments included. So, from this point of view, HU does appear on certain -more formal- occasions in modern Greek.¹ Foundalis: ηυ (eta + upsilon) is extremely rare in Modern Greek; it appears in two verb-forms only: απηυδησα [apivthisa] (=‘I got fed up’), and απηυθυνα [apifthina] (=‘I gave sb. the right to speak’).²

1. Manolis Nikolaou, Eta + Upsilon diphthong, 01 April 2002, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2002-April/020521.html (accessed 03 February 2007).

2. Harry Foundalis, About the Greek Language, http://www.cogsci.indiana.edu/farg/harry/lan/grkphon.htm (accessed 03 February 2007), Phonology and Orthography.

Dentalization of Lamba, Nu and Tau

English l, n, and t are alveolars, that is, pronounced by touching the tip of the tongue to the teethridge. Greek λ, ν, and τ are dentals* – similar to th in this – pronounced by instead touching the tongue to the back of the upper teeth.

To produce Greek [l], [n], and [t/d], lisp English \l\, \n\, \t\, and \d\ by flattening the tongue against the palate and touching the back of the teeth.

Dentalizing [l] and [n] produces the palatalization of λ and ν that occurs before [i] (though makes insignificant difference before non-[i] vowels).

Dentalizing [t] eliminates the aspiration present on English t before stressed vowels, but which does not occur in Greek. (English tot, for example, is identical to \t\ + hot. Dentalizing makes it \t\ + ot.) Dentalizing [d] (when ντ) has insignificant effect other than to slightly soften it.

*A subtle difference between English and Greek pronunciation is that the English consonants ‘l,’ ‘n,’ ‘t,’ and ‘d’ are lingua-alveolar, whereas the corresponding Greek consonants ‘λ,’ ‘ν,’ ‘τ,’ and ‘ντ’ are lingua-dental.¹

1. The Divine Music Project, St. Anthony's Greek Orthodox Monastery, Florence, Arizona, http://www.stanthonysmonastery.org/music/Pronunc.htm (accessed 05 February 2007), Guidelines for Greek Pronunciation.

Zeta and Sigma as ~\zh\ and ~\sh\

When zeta precedes [o], it sounds almost like [ʒ] \zh\ as in pleasure. Similarly, when sigma follows [o], it sounds almost like [ʃ] \sh\ as in sure. These, however, are only auditory illusions created by the Greek [o],* which is not identical to English \oh\.

Foundalis: If you listen carefully to native Greek speakers, it [sigma] sounds a bit between [s] and [sh].... To the native English ear it sounds much closer to [sh] than to [s], whereas every native Greek speaker would swear they pronounce it exactly like the English [s]; hearing [sh] sounds very foreign to the Greek ear.¹

[ʒ] and [ʃ] are produced by puckering the lips. In Greek, when [z] precedes [o] and when [s] follows [o], the lips pucker like [ʒ] and [ʃ] giving them zh- and sh-sounds. This is an effect of the formation of the Greek [o]-vowel rather than a characteristic of the consonants.

To approximate Greek [zo], replace \z\ in the name Zoe with \zh\ to create a puckered \oh\ closer to the Greek [o]. Then combine the \z\ from Zoe with the \oh\ in zhoe to produce a puckered \z\.

διαβάζω [ðʝa-'va-zo] ~\'thya-'va-zoh\ (word does not occur in the GNT)

To approximate Greek [os], append \s\ to \zhoh\, while keeping the \oh\ puckered, to produce a puckered \s\.

γέρος ['ʝe-ros] ~\'yeh-rohs\ (word does not occur in the GNT)

*Sarri: In greek, there is no ʒ postalveolar sound as in french je² and: In greek, there is no ʃ postalveolar sound as in shop.³ Foundalis: There are sounds common in other languages that do not exist in Greek ... ([sh] as in ‘shop’, [Z] as in ‘pleasure’, [ch] as in ‘church’, and [dZ] as in ‘job’).⁴

1. Harry Foundalis, About the Greek Language, http://www.cogsci.indiana.edu/farg/harry/lan/grkphon.htm (accessed 03 February 2007), The Alphabet.

2. Katerina Sarri, Greek Language and Alphabet, http://users.otenet.gr/~bm-celusy/chart2.html#z (accessed 03 February 2007), Table of Greek Consonants and Combinations.

3. Katerina Sarri, Greek Language and Alphabet, http://users.otenet.gr/~bm-celusy/chart2.html#s (accessed 03 February 2007), Table of Greek Consonants and Combinations.

4. Harry Foundalis, About the Greek Language, http://www.cogsci.indiana.edu/farg/harry/lan/grkphon.htm (accessed 03 February 2007), Phonology and Orthography.

Assimilation to Voiced and Voiceless Consonants

In Modern Greek, certain letters assimilate according to whether they precede voiced and voiceless consonants (regressive [or anticipatory] assimilation, in the case of σ, αυ, ευ, and ηυ) or follow (progressive assimilation, in the case of [i]-vowels):

• Sigma is \z\ (voiced) before voiced consonants, and \s\ (voiceless) before voiceless consonants (optionally \s\ or \z\ before λ):
  
  σβέννυμι ~\'zbeh-nyee-mee\ (to extinguish)
  σμύρνα ~\'zmeer-na\ (myrrh)
• αυ, ευ, and ηυ are voiced (~\av\, ~\ehv\ and ~\eev\) before vowels and voiced consonants, and voiceless (~\ahf\, ~\ehf\ and ~\eef\) before voiceless consonants.
• The [i]-vowels (ι, η/ῃ, υ, ει, οι, and υι), when unstressed, palatalize to either [ʝ] or [ɲ] (voiced) between voiced consonants and vowels other than [i]; and palatalize to [ç] (voiceless) between voiceless consonants and vowels other than [i]. (See Palatalization below.)

Voiced consonants are those which vibrate the vocal cords; voiceless consonants are those which do not:

note: Obstruents are characterized by the obstruction of airflow in the vocal tract and form pairs of voiced and voiceless counterparts; the pairs have the same articulative formation, and differ only in the presence or absence of vocal vibration. Sonorants are characterized by relatively free and unobstructed airflow in the vocal tract.

*When β is pronounced [b] (in Classical and Erasmian Greek), it is the voiced counterpart of π [p]; when β is pronounced [v] (in Modern Greek), it is the voiced counterpart of φ [f].

†When γ is pronounced [ɡ] (in Classical and Erasmian Greek), it is the voiced counterpart of κ [k]; when γ is pronounced [ɣ] (in Modern Greek), it is the voiced counterpart of χ [x].

‡When δ is pronounced [d] (in Classical and Erasmian Greek), it is the voiced counterpart of τ [t]; when δ is pronounced [ð] (in Modern Greek), it is the voiced counterpart of θ [θ].

Palatalization

Foundalis: When Greeks want to mockingly imitate the non-native Greek pronunciation of foreigners, the main trick they do is that they avoid palatalization.¹

Palatalization occurs when the middle or back of the tongue touches the roof of the mouth, or hard palate. English has one palatal, y (represented in IPA as [j]). In Greek, certain letters palatalize – acquiring a y-sound (some more acutely than others) – in combination with certain vowels.

The consonants γ, κ, and χ palatalize both before [i] (that is, before ι, η/ῃ, υ, ει, οι, and υι) and before [e] (before ε, αι, and ευ). Λ and ν palatalize before [i] only. The [i]-vowels themselves also palatalize, when followed by a non-[i] vowel and not stressed.

Γ, κ, χ, λ, and ν always palatalize before [i]; γ, κ, and χ always palatalize before [e]:

γυνή [ʝi-'ɲi] ~\yee-'nyee\ (woman, wife)
καί ['ce] ~\'kyeh\ (and, also)
χήρα ['çi-ra] ~\hyee-ra\ (widow)
πολύς [po-'ʎis] ~\poh-'lyees\ (much, many)

[i] also palatalizes if unstressed and followed by another vowel other than another [i] (note, too, that palatalization affects syllabication):

κύριος ['cir-ʝos] ~\'kyeer-yohs\ (lord, sir)
μιᾷ ['mɲa] ~\'mnya\ (inflection of εἷς/μία/ἕν, one)
ἀλήθεια [a-'ʎiθ-ça] ~\a-'lyeeth-hya\ (truth)

[i] is not palatalized whenever it has separate status, either by being followed by a consonant, or by being stressed:

μία ['mi-a] ~\'mee-a\ (one)

When both γ, κ, χ, λ, or ν, and [i] palatalize (that is, when γ, κ, χ, λ, or ν is followed by [i], and [i] is followed by a non-[i] vowel, and [i] is unstressed), the palatalized [i] is subsumed (that is, the [i] is not pronounced):

ἅγιος ['a-ʝos] ~\'a-yohs\(holy)
εὐαγγέλιον [ev-aŋ-'ɟe-ʎon] ~\ehv-ang-'gyeh-lyohn\ (good message)
ἀσθένεια [as-'the-ɲa] ~\as-'theh-nya\ (weakness)

Γ, κ, and χ do not subsume [e] because [e] never palatalizes:

δίκαιος ['ði-ce-os] ~\'thee-kyeh-ohs\ (righteous, just)

When [i] begins a word – and is unstressed and followed by a non-[i] vowel...?

ἰῶτα ['ʝo-ta] ~\'yoh-ta\ (iota)
ἰάομαι [i-'a-ome] ~\ee-'a-oh-meh\ (to heal)
υἱός [i-'os] ~\ee-'ohs\ (son)

h. [ʝ]and [ɟ]: [ʝ] and [ɟ] are both very nearly like [j], the \y\ sound in yes. Consult the IPA Sound Chart for the difference between [ʝ], [ɟ] and [j].

i. [c]:

j. [ç]:

k. [ʎ]:

l. [ɲ]:

m. [ʝ]: [ʝ] is a voiced consonant, as are β, δ and ζ (and also γ, λ and ν).

n. [ɲ]: [ɲ] is a voiced consonant, as is μ.

o. [ç]: [ç] is a voiceless consonant, as are θ, π, ξ, σ, τ, φ and ψ (and also κ and χ).

1. Harry Foundalis, About the Greek Language, http://www.cogsci.indiana.edu/farg/harry/lan/grphdetl.htm#p_palatalization (accessed 03 February 2007), Palatalization.

Notes on Koine

a. αυ, ευ and ηυ: Smyth: After υ had become like Germ. ü [fifth century BCE], the only means to represent the sound of the old υ (oo in moon) was ου. Observe, however, that in diphthongs, final υ retained the old υ sound.¹

b. αυ, ευ: Buth cites Horrocks that αυ and ευ began transitioning in 3^rdc. BCE, to [aβ] and [eβ] by the Roman period, and perhaps already Modern [ɑv]/[ɑf] and [ev]/[ef], such that υ was increasingly being interchanged with β in the late Roman/early Byzantine period.²

c. αυ: Buth cites a 2^ndc. CE papyrus showing αυ for α (ΦΛΑΥΟΥΙΟΥ for Φλαουίου),³ indicating (only the changing nature of υ?).

♦ αυ, ευ: Buth cites a Ptolemaic [281-30 BCE] papyrus showing αυ for αβ (ραυδους for ῥάβδους), and a 1^stc. CE papyrus showing εβ for ευ (Πνεβτῦνι for Πνευτῦνι)⁴, exhibiting a Modern (or proto-Modern) pronunciation in which αυ/ευ is either [av]/[ev] or [aβ][eβ] before both voiced and voiceless consonants.

e. ηυ: Mastronarde: It is very hard for English-speakers to distinguish between ηυ and ευ, and the distinction was lost in Attic itself in the course of the fourth century, so that words that properly had ηυ appeared spelled with ευ.⁵

f. β: Buth: Already at the beginning of the Roman period [146 BCE] βῆτα was becoming a ‘soft’ bilabial fricative, probably like Spanish ‘b’ in Havana/Habana (the city).⁶

g. β: The traditional Greek transliteration of Classical Latin v (pronounced \w\ up until the 1^stc. CE) was ου. Conrad (B-Greek, 1996): We find historians writing Greek in the first and later centuries of our era transliterating Roman names like Varus and Vergilius as OUAROS and OUERGILIOS, which should mean that the OU + vowel was our W sound.⁷

h. β: Carlson (B-Greek, 1996): The fricative pronunciation of Latin V first appeared sporadically in the first century A.D., became wide-spread in the second century, but with a few pockets of the former pronunciation (‘w’) surviving until the fifth century.⁸

♦ β: Coincident with the Latin shift from \w\ to \v\ is a Greek shift in the transliteration of Latin v. Buth cites 1^st and 2^ndc. CE examples of β for Latin v, including a 2^ndc. CE papyrus showing Latin Silvanus transliterated first as Σιλβανου and later corrected to Σιλουανου.⁹ The question is whether this informs us about the pronunciation of β. If Greek β were still \b\, would it, or would ου, be the natural choice for translitering the new Latin \v\?
IABE Phonology 101

♦ γ: Buth cites a 1^stc. BCE example of ιγε for ιε, and two 1^stc. CE examples of υγι for υι, and ι for γ,¹⁰ evidencing palatalization of γ as [j] \y\. Buth: The insertions and the substitutions with ι would not be probable without γάμμα having become a soft fricative.¹¹

k. δ: Buth cites Gignac: According to Gignac ... δελτα first became fricativized before ι, around the first century. Interchanges between δ and ζ ... only begin from the third century ... evidenc[ing] the complete fricativization of δελτα.¹²

l. χ and φ: Buth quotes Horrocks: There is also possible evidence for a fricative pronunciation of /k^h/ (second century BC) and /p^h/ (second century AD) in the Asia Minor Koine.... Though the evidence is frankly meagre, it would perhaps be reasonable to assume that frication in the Koine began in various areas outside Egypt during the Hellenistic period and that it had been widely, though by no means universally, carried through by the end of the fourth century AD.¹³

♦ φ: Buth: Most of our colloquial papyri come from Egypt where the local Coptic seems to have encouraged a hard p^h, t^h, k^h. With the voiceless fricatives we have Attic inscriptions with some evidence of soft forms already in the second century CE.¹⁴ Buth subsequently cites a 2^ndc. CE Attic (i.e., northern) inscription substituting εφ for ευφ,¹⁵ exhibiting a Modern (or proto-Modern) pronunciation in which ευ is either [ef] or [eɸ] before a voiceless consonant (such that ευφ acts as a double consonant: εφφ).

♦ θ: Buth cites a 2^ndc. CE Judean (i.e., southern) papyrus showing θ for τ,¹⁶ evidencing the Classical pronunciation [tʰ] of θ.

o. η: Buth: The vowel η became like ι and ει by the fourth century CE.... Gignac is of the opinion that η merged with ι in the second century CE.... Broadly speaking, it would appear that most people ... used it as an equivalent for a close/mid-high ... sound [between ι/ει and ε/αι] in the early Roman period.¹⁷

p. υ, οι: Smyth: υ was originally sounded as u in prune, but by the fifth century [BCE] had become like that of Fr. tu, Germ. thür.¹⁸

q. υ, οι: Sarri: In ancient Attic the Upsilon was pronounced [y] as in french tu.... The /y/ pronunciation did survive till the 9/10th century C.E. But, after the 3rd century C.E. the Y and OI pronunciation coincided within the iotacistic tendency of turning many sounds to [i]. After the middle times, Upsilon was pronounced [i]. ... The Romans called it y Graecum since they had no sound [y] (their U was pronounced [u]).¹⁹

♦ υ, οι: Buth cites a 2^ndc. CE papyrus showing υι for υ (ενγυιου for εγγυου): an extra ι ... probably reflecting a front vocalic [ü] pronunciation of υ.²⁰ (Note also the pronunciation of γγ as \ng\.)

♦ υι: Buth cites a 100 CE papyrus which, by substituting οιει for υι (οιειωι for υιῷ), shows οι for υ, and ει for ι and indicates a non-diphthongal pronunciation of υι.²¹ Buth (B-Greek, 2000): I keep UI distinct, on the basis of a 1st century papyrus that spells ‘to the son’ [υἱῷ] as OIEIW OI and EI appear to have separate status.²²

1. Herbert Weir Smyth, A Greek Grammar for Colleges (New York [etc.]: American Book Company, 1920), available online at Christian Classics Ethereal Library: http://www.ccel.org/s/smyth/grammar/png/0027=13.htm, 13.

2. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 9, 179.

3. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 8, 179.

4. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 9, 179.

5. Donald J. Mastronarde, Ancient Greek Tutorials, Berkeley Language Center of the University of California, Berkeley, http://socrates.berkeley.edu/~ancgreek/pronunchtml/sigmaU.html (accessed 03 February 2007), Eta.

6. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 179.

7. Carl W. Conrad, Re: Upsilon, 02 February 1996, B-Greek Mailing List Archives, http://www.ibiblio.org/bgreek/test-archives/html4/1996-02/12176.html (accessed 14 February 2007).

8. Stephen C. Carlson, Re: Upsilon, 02 February 1996, B-Greek Mailing List Archives, http://www.ibiblio.org/bgreek/test-archives/html4/1996-02/12179.html (accessed 14 February 2007).

9. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 179.

10. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 179-180.

11. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 179.

12. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 10, 179.

13. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Footnote 13, 180.

14. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Discussion of Consonants, 180.

15. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Insciption Graeca 2.11507, 180.

16. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), BenKosiba 1.8, 180.

17. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), The Satus of η, 178.

18. Herbert Weir Smyth, A Greek Grammar for Colleges (New York [etc.]: American Book Company, 1920), available online at Christian Classics Ethereal Library: http://www.ccel.org/s/smyth/grammar/png/0027=13.htm, 13.

19. Katerina Sarri, Greek Language and Alphabet, http://users.otenet.gr/~bm-celusy/upsilon.html (accessed 14 February 2007), Full Upsilon Discussion.

20. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Babatha 21.17,25, 178.

21. Randall Buth, Notes on the Pronunciation System of Phonemic Koine Greek (PDF), The Biblical Language Center, http://www.biblicalulpan.org/pages/2006/greekssmat1.html#Pro_notes (accessed 03 February 2007), Papyrus 109.2, 177.

22. Randall Buth, Modern Greek, ... a new direction-Peterson, 18 December 2000, B-Greek Mailing List Archives, http://lists.ibiblio.org/pipermail/b-greek/2000-December/014673.html (accessed 14 February 2007).

Long and Short Vowel/Checked and Free Vowels

Greek maintains a grammatical distinction between long and short vowels, though all have been pronounced short since the early Koine period.

In certain contractions and in certain verb tenses, short vowels lengthen according to the following rules:

(partial:)
ε + α = η

Audio Resources

A sample reading of Mark 2 in Modern pronunciation from Koine Greek New Testament: Narrated by Dr. Spiros Zodhiates is available from AMG Publishers.

A sample reading of 1 John 1 in Living Koine from LIVING KOINE GREEK for EVERYONE is available from Dr. Buth's Biblical Language Center.

Learn Greek through the Internet

Filoglossia

References