Greek Pronunciation

Homepage | Index | Mythology | Radio | Music | Dances | Northampton

The “Art of Grammar” written in about 100 BC by Dionysios Thraikos shows that ancient Greek was pronounced in exactly the same way as modern Greek because of the way he describes how the sounds of the Greek letters are constructed.

According to Dionysios Thraikos the ancient Greek alphabet contains 24 letters, 7 of which, α, ε, η, ι, ο, υ, ω are called Φωνήεντα or Vowels because they are voiced of their own accord.

Two vowels are called Μακρά or Long η and ω, two vowels are called Βραχέα or Short ε and ο, and three are called Δίχρονα or Double α, ι, υ because they can either be stretched out or shortened. There are five προτακτικά or prefix vowels α, ε, η, ο, ω and two υποτακτικά or suffix vowels ι, and υ whereby υ only becomes a prefix in front of ι. Six Δίφθογγοι or Diphthongs also exist, αι, αυ, ει, ευ, οι and ου. (Note that αε, οε, ια, υι, and other similar combinations are not and have never been classified as diphthongs.)

There are 17 Σύμφονα or Consonants β γ δ ζ θ κ λ μ ν ξ π ρ σ τ φ χ ψ which are so called because they do not have a voice by themselves. Eight of these are called ημίφωνα or semi-voiced, ζ ξ ψ λ μ ν ρ σ because they are voiced due to moaning or hissing. Nine Άφωνα non-voiced consonants exist, β γ δ κ π τ θ φ χ, which are so called because they are exceedingly more unnatural than the others and their song is unnatural.

Dionysios Thraikos continues by saying that the non-voiced consonants form into three groups. Three, κ, π, τ are called Ψιλά or Bare. Three, θ, φ, χ are called Δασέα or Rough and three are called Μέσα Δε Τούτων or Middle Of These, β, γ, δ. These sounds are called Middle because their sound is the middle value of the combination of two letters from each of the previous groups so that β is in-between of π and φ, γ is in-between κ and χ and δ is in-between of θ and τ.

Finally there are three letters ζ, ξ, ψ called Διπλά or Double because they are made up of two consonants, thus ζ is made up of σ and δ, ξ is made up of κ and σ and ψ is made up of π and σ.

From the above description it is clear that ancient Greek could only have been pronounced in the same way as modern Greek otherwise the combinations shown by Dionysios Thraikos do not make sense.

According to Grimm’s Law the hard consonantal sounds in all Germanic languages evolved from softer sounds which Grimm called “Voiced Aspirated” consonants. According to Grimm the “Voiced Aspirated” consonants changed into so-called “Voiced” consonants and the “Voiced” consonants then became so-called “Voiceless” consonants in Germanic languages. The “Voiced Aspirated” term which Grimm used was entity arbitrary and had no scientific meaning but Grimm decided that the “Voiced Aspirated”, labial, dental and velar sounds should be represented as bh, dh, and gh. It was up to individual linguists to decide what these should sound like but based on the description given by Dionysios Thraikos it is obvious that these are the sounds of the Greek letters beta, delta and gamma and this is why these ancient Greek letters were used in early Germanic Gothic script to represent the letters b, d and g which still retained their “Voiced Aspirated” sounds at the time this script was devised.

The following table can be constructed for English based on Grimm’s Law to make things easier to understand. Note that Labial sounds are made with the lips, Dentals with the tongue touching the teeth and Velars with the tongue at the back of the month. Aspirated sounds are made with a puff of breath. Grimm’s “Voiced” consonants can only be made if they are preceded or followed by a vowel and are thus the same as Dionysios Thraikos non-voiced consonants. Grimm’s “Voiceless” consonants can be spoken without use of a vowel if they are combined with s and are therefore the same as Dionysios Thraikos semi-voiced consonants. Grimm’s terminology differs from that of Dionysios Thraikos because by “Voiced” and “Voiceless” Grimm is referring to the requirement of a vowel to be in combination with the consonant to produce the sound of the consonant for “Voiced” and the non-requirement of a vowel to sound the consonant for “Voiceless” consonants whereas Dionysios Thraikos uses the term voiced to mean a pure vowel, con-voiced to mean a consonant and the terms non-voiced and semi-voiced to mean non-vowel and semi-vowel. Aristotelis uses exactly the same terminology as Dionysios Thraikos in part XX of his “Poetics” published in 350 BC which deals with Greek grammar.

English	labial	dental	velar
Voiceless	p	t	k
Voiced	b	d	g
Voiced Aspirated	bh	dh	gh

English	labial	dental	velar
Aspirated	f	th	h

The basic principle of Grimm’s law applies to all Indo-European languages so for Greek both ancient and modern the same table can be written as follows using Dionysios Thraikos linguistic terminology.

Greek	labial	dental	velar
Semi-Voiced	Ψ-Σ		Ξ-Σ
Bare	Π	Τ	Κ
Middle	Β	Δ	Γ

Where Ψ-Σ indicates the sound of psi minus the sound of sigma, etc.

Greek	labial	dental	velar
Rough	Φ	Θ	Χ

The above tables show that Dionysios Thrakos description of the way the sounds of beta, delta and gamma are made corresponds to the “Voiced Aspirated” group of Grimm.

The next table equates the terminology used by Grimm with that used by Dionysios Thraikos and Aristotelis (“Poetics XX”).

Linguiststerms	=	=	=	=	=	=
Grimm	Voiceless	Voiced	Voiced Aspirated	Aspirated	Consonant	Vowel
Thrakos	Semi-voiced	Bare	Middle	Rough	Con-voiced	Voiced
Aristotelis	Semi-voiced	Non-voiced	Non-voiced	Non-voiced	Con-voiced	Voiced

Note that in English p, t and k tend to be aspirated whereas in most other indo-European languages they are un-aspirated and therefore sound closer to b, d, and g. Aspirated p is p as in print and un-aspirated p is p as in spin, sprain or suspend. Un-aspirated t is t as in stop, strain or register. Un-aspirated k is k as in sky, scrape or whisker. 

Pronunciation of the Greek alphabet in English

Pronunciation of the English alphabet in Greek

Αα= a as in cat. Ββ= v as in visit Γγ = γin γάλαnonexistent in English, or y in γυναίκα Δδ= th in this Εε  = high e as in elf (F f = w ancient Greek digamma) Ζζ= z as in zoo Ηη= long high i as in tin Θθ= th as throat Ιι= ee as in meet Κκ= g as in gap Λλ= l as in log Μμ= m as in mat Νν= n as in not Ξξ= ks as in axe Οο= short o as in cot Ππ= b as in bit (Q q = q ancient Greek qoppa) Ρρ= r as in rot Σσς= s as in sing Ττ= d as in dart Υυ= high y as in cheat Φφ= f as in fish Χχ= h as in heat Ψψ= ps as in lapse Ωω= long o as in store αι = e as in bet αυ= av and in have or af as is after ει= i as in pain ευ= ev and in ever or ef as in effect οι= i as in coin ου= oo as in cook

A a = α in bag or εΐ in ape B b = π C c = σin cyan or k in cattle D d = τ E e = ι in English or ε in egg F f = φ G g = κ H h = χ I i = αΐ in light or ι fit J j = Cypriot κ in και K k = the first sound of Ξ Ll = λ Mm = μ Nn = ν Oo = ο P p = the first sound of Ψ Q q = q ancient Greek qoppa R r = ρ S s = σ T t = sound nonexistent in Greek U u = sound nonexistent in Greek V v = β W w = f ancient Greek digamma ? X x = ξ Y y = γas in γυναίκα Z z = ζ sh = Cypriot σσin Βίσση ch = nonexistent in Greek th = θ the = δ ps = ψ ks = ξ

Because non-Greek (mostly English and Germanic) linguists used a completely synthetic artificial pronunciation of Greek since the time of Erasmus based on the native pronunciation of the languages of their immediate neighbours and deliberately ignored the native pronunciation of Greek when they formulated their theories concerning the origin of indo-European languages in the 19^th century they made fundamental errors in reconstructing ancient Greek pronunciation and the way the so-called proto-indo-European root language was pronounced.

An analysis of ancient Greek texts by Dr. Chrys C. Caragounis has proven that ancient Greek was pronounced almost exactly like modern Greek since classical times and that by the time of Christ all of the vowels and consonants except for ita (η) were pronounced the same way as today by the majority of native Greek speakers. Overturning the claims of non-Greek linguists Dr. Caragounis has shown that ancient Greek words would be unpronounceable if the consonants β, δ, γ, φ, θ, χ were pronounced b, d, g, p^h, t^h, k^h as Erasmians claim.

The Dactylic Hexameter used by Homer also shows that ancient Greek could not have possibly been pronounced using the Erasmian or reconstructed pronunciation systems since it would be out of meter.

Dactylic Hexameter adheres to the following pattern of syllables

- U | - U | - U | - U | - u u | - -

where - represents a long syllable, u represents a short syllable and U is either one long or two short syllables.

A combination of a long syllable followed by two short syllables is called a dactyl and a combination of two long syllables is called a spondee.

The following verse from Homer consists of dactyl, dactyl, spondee, dactyl, dactyl, spondee.

Μήνιν ά είδε, θεά, Πηληΐάδεω Αχιλήος

Μήνιν ά | είδε, θε | ά, Πηληΐά | δεω Αχι | λήος

mhnin a | eide, the | a, Ph | lh.i.a |  dew Axi | lhos

If the Barbaric pronunciation of ancient Greek were used then instead of “eide the” being one long and two short syllables “ei-de the” as the meter requires and which is given by the Greek pronunciation, you would have “e-i-de t-He” or 5 short syllables with ei being pronounced as two separate sounds “e” and “i” with the stress on the “i” (instead of a single diphthong with one sound), and “th” being pronounced as two separate letters “t” and “h” (instead of one single sound), which renders what should be a dactyl into something which is not and destroys the entire meter.

Even if we assume that the Barbaric pronunciation had been the original pronunciation of Greek in Mycenaean’s times when Greek started being written in Linear B in about 1600 BC, it is clear given Dionysios Thraikos account showing modern pronunciation was fully formed in 100 BC, that the point at which the Greek pronunciation become dominant over Barbaric pronunciation with the majority of the population using it was (1600+100)/2 = 850 BC which is around about the time of Homer.

The Barbaric pronunciation of proto-indo-European, Sanskrit, Greek and Latin is shown below

π

τ

κ

κ^w

β

δ

γ

γ^w

φ

θ

χ

χ^w

Proto-Indo-European

p

t

k

k^w

b

d

g

g^w

b^h

d^h

g^h

g^hw

Sanskrit

p

t

s

k/s

b

d

j

g/j

b^h

d^h

h

g^h/h

Ancient Greek

p

t

k

p/t/k

b

d

g

b/d/g

p^h

t^h

k^h

p^h/t^h/k^h

Modern Greek

p

t

k

p/t/k

v

dh

gh

b/d/g

f

th

h

f/th/h

Latin

p

t

c

qu

b

d

c

v

f

f

h

f

Old English

f

th

h

hw

p'

t'

c

cw/k

b

d

g

b/g

The Greek pronunciation of Greek (ancient and modern) is shown here

π

τ

κ

κ^w

β

δ

γ

γ^w

φ

Θ

χ

χ^w

Greek

b'

d'

g’/j

b/d/g

v

dh

gh

v/dh/gh

f

th

h

f/th/h

Note that p, t and k are un-aspirated in all languages except English. Un-aspirated p/t/k are virtually identical to b’/d’/d’.

Art of Grammar by Dionysios Thraikos

The Error of Erasmus and Un-Greek Pronunciations of Greek

History of the Mycenaean’s

Linear-B inscriptions

Gods named in Linear-B

According to my Web-Counter you are visitor Number

since 1st August 1996

 last update 21-05-2008