In this episode of the Koine Greek video blog, I deal with the pronunciation of voiced stops following nasals. Some helpful vocabulary for listening to the video:

σύμφωνον = consonant

ἔρρινον = nasal (nasal)

ἔκκροτον = plosive/stop (neologism)

τριβόμενον = fricative (neologism)

ὀρθογραφία = orthography (neologism)

I must admit to a fair number of mistakes in this video and some clumsy explanations. It was quite a difficult subject to address in Koine Greek, but I did my best! I will hopefully make more such videos and improve every time. That's what this is all about!

However, though it is in video form above, because of the interest in such a topic, I have decided to write out the main points of the blog in English as well below (the Egyptian material below is from Gignac 1976 with some comments from Petrounias 2007 and Horrocks 2014 on general Koine; the Palestinian Koine examples are from Kantor 2017; Modern Greek material is from Holton et al. 2012 and Λεξιξο της Κοινης Νεοελληνικης here ):

Voiced Stops: β , δ , γ

In Classical Greek, the voiced stops β, δ, γ were pronounced as [b], [d], [g], respectively. That means that Greek β was pronounced as the b in the English word boy. Greek δ was pronounced as the d in the English word do. Greek γ was pronounced as the g in the English word go:

*Classical Greek*
β	λάβε 'take!'	[ˈlabe]
γ	γάμος 'write!'	[ˈgamos]
δ	δός 'give!'	[ˈdos]

During the Koine period, however, these sounds were gradually becoming fricatives.

Greek γ

Greek γ was coming to be pronounced as a voiced velar fricative (IPA: [ɣ]) in most environments and as a palatal approximant (IPA: [j]) in the environment of high front vowels. In non-linguistic terms, this means that usually it is pronounced like a "gargling" sound (like a cat purring), but when you say it around the vowels i and e it sounds like English y in the word year. This is evidenced by the following spelling interchanges:

*Egyptian Koine Greek*
ἱγεροῦ (for ἱεροῦ)	5/4 BCE
εωργῶι (for γεωργῶ)	3 CE
ὑιῆ (for ὑγιῆ)	137 CE

*Palestinian Koine Greek*
αγω (for ἁγίῳ)	605-606 CE

One piece of evidence for Palestinian Koine may be the fact that the emperor Trajan's name, Trajanus (IPA: [trajanus]) in Latin, is rendered as טרוגינוס ( trwgynws) in the Mekhilta DeRabbi Shimon Bar Yochai (21:13) and as טרוגיינוס ( trwgyynws) in the Jerusalem Talmud (Sukkah 23a). It is possible that this reflects the Greek variant spelling Τραγιανος which is also attested in Egyptian papyri. On the other hand, the normal spelling of Trajanus in Palestinian Koine Greek is Τραιανος.

It is likely that this change happened still during the Ptolemaic period (at least as is evidenced from Egyptian Koine) by the mid-second century BCE.

Greek β

Greek β was coming to be pronounced as a voiced bilabial fricative (IPA: [β]) in most environments. This means that Greek β was becoming like the Spanish b/v sound: e.g., trabajar 'to work' (IPA: [t̪ɾaβaˈxaɾ]). This is evidenced by the following spelling interchanges:

*Egyptian Koine Greek*
ἕδομον (for ἕβδομον)	47 CE
Πνευτύνις (for Πνεβτύνι)	35/36 CE

*Palestinian Koine Greek*
Βερουταριου (for Latin Verutarius)	1^st c. BCE/1^st c. CE
Σεβηρα (for Latin Severus)	3^rd-6^th c. CE

This change had likely occurred by the first century CE in both Egypt and Palestine.

Greek δ

Greek δ was coming to be pronounced as a voiced dental fricative (IPA: [ð]). This means that it sounds like the th sound in the English word this. Such a change is evidenced by the following spelling interchanges:

*Egyptian Koine Greek*
δαπέσηι (for διαπέσηι)	3 CE
δώρυγος (for διώρυγος)	110 CE
δά (for διά)	208 CE
ζώρυγα (for διώρυγα)	3^rd c. CE
ζακοσίας (for διακοσίας)	268/269 CE
σπούζασων (for σπούδασον)	4^th c. CE
ζώτε (for δότε)	5^th/6^th c. CE

It is possible that this change began before the vowel [i] already in the first century CE and was completed in all other environments by the third or fourth century CE (Gignac's view). It is more likely, however, that the change did not actually occur until the 3^rd/4^th century CE (Horrocks's view).

Pronunciation of the Voiced Stops Before Nasals

While many who are familiar with the "Living Koine" pronunciation-I use it on this website-are aware of these facts, I believe there is less awareness regarding the realization of these consonants after nasals. By nasals I mean those sounds you make where air comes out of your nose, namely, μ [m] and ν [n].

Even though these changes took place in the Koine period and the "fricative" realizations [ɣ]/[j], [β], [ð] became the regular pronunciation of γ, β, δ, respectively, they maintained their "stop" pronunciations of [g], [b], [d] when they followed the nasal consonants μ [m] and ν [n].

There are two main pieces of evidence for this.

First, the following spelling interchanges suggest that the "stop" pronunciation was maintained following the nasal consonants even after the fricative pronunciation had taken over elsewhere (note the dates in comparison with the suggested dates for the fricative pronunciations above):

*Egyptian Koine Greek:* γ *following a nasal*
ἐνκύς (for ἐγγύς)	unknown
συνγεχοριγένε (for συγκεχωρηκέναι)	46 CE
ἀγγάλαις (for ἀγκάλαις)	592 CE

*Egyptian Koine Greek:* β *following a nasal*
γαμρᾶς (for γαμβρᾶς)	45-47 CE
γαμροῖς (for γαμβροῖς)	2^nd/3^rd c. CE
ἐμπῆναι (for ἐμβῆναι)	35 CE
γαμπρά (for γαμβρά)	172-173 CE
παρεμπολήν (for παρεμβολήν)	3^rd c. CE
πέμβτης (for πέμπτης)	33 CE
σὺμ βρονησίῳ (for σὺν προνησίῳ)	3^rd c. CE

*Egyptian Koine Greek:* δ *following a nasal*
τήντε (for τήνδε)	30/31 CE
ἀντρός (for ἀνδρός)	32 CE
κίντυνων (for κίνδυνον)	167/168 CE
ὑπάρχονδα (for ὑπάρχοντα)	32 CE
πένδε (for πέντε)	45 CE

Second, though the fricative pronunciation of γ, β, δ has been generalized across the board in Modern Greek as the fricatives [ɣ]/[j], [v], [ð] (e.g., λαμβάνω [lamˈvano] and not [lamˈbano] as in Koine), there are still hints of a distinction between the sounds indicated by these letters when following nasals in some specific words that go back to the Koine period. If we trace some Classical/Koine words into Modern Greek, we find that this distinction is occasionally preserved even in the spelling (note that Modern Greek uses γγ or γκ for [g], μπ for [b], and ντ for [d]):

*Modern Greek:* γ
αγγούρι [a(ŋ)ˈguri] 'cucumber'	cf. Class. ἀγγ ούριον (Koine: [a ŋˈguri(o)n])
αγόρι [aˈ ɣori] 'boy'	cf. Class. ἄγ ουρος (Koine: [ˈa ɣuros])

*Modern Greek:* β
αμπολή [am boˈli] 'water barrier'	cf. Class. ἐμβ ολή (Koine: [εmb oˈle])
αναβολή [anavoˈli] 'postponement'	cf. Class. ἀναβ ολή (Koine: [anaβ oˈle])

*Modern Greek:* δ
δέντρο [ˈðendro] 'tree'	cf. Class. δένδ ρον (Koine: [ˈðεn dron])
δίεδρος [ˈðieðros] 'dihedral'	cf. Class. δίεδ ρος (Koine: [ˈðie ðros])

* The final example with τ for δ in δεντρο may simply be due to the ρ.

Conclusions

Therefore, in light of the pronunciation and pedagogy principles outlined by Buth, especially #1 and #3 (see here ), we might want to consider adopting this pronunciation in our speech as well. On the other hand, it would conflict a bit with #4, inasmuch as Modern Greek generalizes the fricatives basically across the board.

Nevertheless, adopting such a pronunciation may even help readers of ancient papyri and manuscripts more readily discern confusion between certain words that would have been pronounced the same in the Koine period, even though they would be spelled differently. For example, in a text without word divisions, it is entirely possible that the sequence φιλωτηντε πολινκαιτοεθνοσ could reflect either of the following normalizations:

φιλῶ τήνδε πόλιν καὶ τὸ ἔθνος 'I love this (dem.fsg.acc) city and the nation'

φιλῶ τήν τε πόλιν καὶ τὸ ἔθνος 'I love both (conj) the (art.fsg.acc) city and the nation'

Therefore, adopting such a pronunciation is both consistent with the historical pronunciation as well as the trajectory of the language and important for understanding the range of possible meanings in some instances.

To put it simply, in Living Koine, we might consider pronouncing the voiced stops as fricatives generally, but as stops after nasals:

*Living Koine Pronunciation*
γάμος 'wedding'	[ˈɣamos]
ἄγγελος 'messenger'	[ˈaŋgεlos]
λάβε 'take!'	[ˈlaβε]
λαμβάνω 'I take'	[lamˈbano]
δός 'give!'	[ˈðos]
ἀνδρός 'man.gen'	[anˈdros]

Gignac, Francis Thomas. 1976. A Grammar of the Greek Papyri of the Roman and Byzantine Periods: Volume I: Phonology . Milan: Istituto Editoriale Cisalpino - La Goliardica.

Holton, David, Peter Mackridge, and Irene Philippaki-Warburton. 2012. Greek: A Comprehensive Grammar.

Horrocks, Geoffrey. 2014. Greek: A History of the Language and Its Speakers. Oxford: Wiley-Blackwell.

Kantor, Benjamin. 2017. "The Second Column (Secunda) of Origen's Hexapla in Light of Greek Pronunciation." diss., The University of Texas at Austin.

Petrounias, E. B. 2007. "Development in Pronunciation During the Hellenistic Period." In A History of Ancient Greek: From the Beginnings to Late Antiquity, edited by A.-F. Christidis, 599-609. Cambridge: Cambridge University Press.

KOINE GREEK VIDEO BLOG #4: Pronunciation of the Voiced Stops β δ γ following Nasals in Koine Greek