Greek orthography

The orthography of the Greek language ultimately has its roots in the adoption of the Greek alphabet in the 9th century BC. Some time prior to that, one early form of Greek, Mycenaean, was written in Linear B, although there was a lapse of several centuries (the Greek Dark Ages) between the time Mycenaean stopped being written and the time when the Greek alphabet came into use.

Early Greek writing in the Greek alphabet was phonemic, and different in each dialect. Since the adoption of the Ionic variant for Attic in 403 BC, however, Greek orthography has been largely conservative and historical.

Given the phonetic development of Greek, especially in the Hellenistic period, certain modern vowel phonemes have multiple orthographic realizations:

• /i/ can be spelled η, ι, υ, ει, οι, or υι (see Iotacism);
• /e/ can be spelled either ε or αι;
• /o/ can be spelled either ο or ω.

This affects not only lexical items but also inflectional affixes, so correct orthography requires mastery of formal grammar, e.g. η καλή /i kaˈli/ 'the good one (fem. sing.)' vs. οι καλοί /i kaˈli/ 'the good ones (masc. pl.)'; καλώ /kaˈlo/ 'I call' vs. καλό /kaˈlo/ 'good (neut. sing.)'.

Similarly, the orthography preserves ancient doubled consonants, though these are now pronounced the same as single consonants, except in Cypriot Greek.

Digraphs and diphthongs

A digraph is a pair of letters used to write one sound or a combination of sounds that does not correspond to the written letters in sequence. The orthography of Greek includes several digraphs, including various pairs of vowel letters that used to be pronounced as diphthongs but have been shortened to monophthongs in pronunciation. Many of these are characteristic developments of modern Greek, but some were already present in Classical Greek. None of them is regarded as a letter of the alphabet.

During the Byzantine period, it became customary to write the silent iota in digraphs as an iota subscript.

Letters	Transliteration		Pronunciation
	Ancient Greek	Modern Greek	Classical Ancient Greek	Modern Greek
αι, αι	ai	e	[ai̯]	[e̞]
αι, ᾳ	āi	a	[aːi̯]	[a]
ει, ει	ei	i	[eː]	[i]
ηι, ῃ	ēi		[ɛːi̯]
οι, οι	oi		[oi̯]
υι, υι	ui		[yː]*
ωι, ῳ	ōi	o	[ɔːi̯]	[o̞]
αυ, αυ	au	av, af	[au̯]	[av] before vowel or voiced consonant; [af] otherwise
αυ, αυ	āu		[aːu̯]
ευ, ευ	eu	ev, ef	[eu̯]	[ev] before vowel or voiced consonant; [ef] otherwise
ηυ, ηυ	ēu	iv, if	[ɛːu̯]	[iv] before vowel or voiced consonant; [if] otherwise
ου, ου	ou	u	[uː] earlier [oː]	[u]
ωυ, ωυ	ōü	oi	[ɔː.u]**	[oi]
γγ, γγ	ng	ng, ny, g, y, ngh	[ŋɡ]	[ŋɡ] in formal registers, but often reduced to [ɡ] in informal speech; also pronounced [ŋɣ] in some words (e.g. εγγενής, έγγραφο, συγγραφέας)***
γκ, γκ	nk	g, y, ng, ny	[ŋk]	[ɡ] word-initially and in some loanwords; [ŋɡ] otherwise, often reduced to [ɡ] in informal speech***
γξ, γξ	nx	nx	[ŋks]	[ŋks]
γχ, γχ	nch	nch, nkh	[ŋkʰ]	[ŋx]***
μπ, μπ	mp	b, mb	[mp]	[b] word-initially and in some loanwords; [mb] otherwise, often reduced to [b] in informal speech
ντ, ντ	nt	d, nd	[nt]	[d] word-initially and in some loanwords; [nd] otherwise, often reduced to [d] in informal speech

* The diphthong υι [yi̯] was monophthongized to [yː] in Classical Attic Greek, but survives in some other contemporary dialects and in early Koine.

** The diphthong ωυ [ɔːu̯] was found in Ionic and in certain Hebrew transcriptions in the Greek Bible, but it did not occur in Attic, and was gradually lost in Koine. Where ωυ was atticized, it was often split into two separate syllables [ɔː.y], hence the Latin transcription ōy. Perhaps the clearest example of this is the Biblical Greek name Μωυσῆς [mɔːu̯.sɛ̂ːs] (Moses), which was atticized as Μωϋσῆς [mɔː.y.sɛ̂ːs], then adapted to early Christian Latin as Mōysēs, from where it became Spanish Moisés, French Moïse, etc. The modern Greek form is Μωυσής [mo̞i̯ˈsis], whereas the modern Latin Vulgate form is Mōsēs.

***The velars [ɡ], [k], [ɣ], and [x] are palatalized to [ɟ], [c], [ʝ] and [ç] respectively before the close and mid front vowels [i] and [e̞].

It is discussed among scholars whether the velar nasal [ŋ] (ἄγμα, ágma) should be regarded as an allophone of /n/ or a phoneme in its own right in Greek.

Diacritics

Main article: Greek diacritics

Polytonic spelling uses a variety of diacritics to represent aspects of the pronunciation of ancient Greek. Polytonic, along with lowercase letters, became standard in Byzantine Greek, although the ancient distinctions had disappeared, replaced by a simple stress accent. The orthographies of modern Greek, both katharevousa and dhimotiki, used the polytonic system until 1982, when monotonic spelling was introduced. In some conservative contexts, such as the Church, polytonic spellings are still used.

Monotonic orthography, adopted in 1982, replaces the ancient diacritics with just two: the acute accent (tónos, e.g. ί), used to mark the stressed syllable in polysyllabic words, and the diaeresis (dialytiká, e.g. ϊ), which indicates that the vowel is not part of a digraph.

Punctuation

Ancient Greek was written scripta continua without spacing or interpuncts. Over time, a variety of symbols appeared. A system of dots credited to Aristophanes of Byzantium was developed in the 3rd century BC: a low dot ⟨.⟩ marked an occasion for a short breath after a short phrase, a middot ⟨·⟩ marked an occasion for a longer breath after a longer passage, and a high dot ⟨˙⟩ marked a full stop at the end of a completed thought. Other writers employed two dot punctuation ⟨⁚⟩ to mark the ends of sentences or changing speakers. Less often, arrangements of three ⟨⁝⟩, four ⟨⁞⟩ and ⟨⁘⟩, and five dots ⟨⁙⟩ appeared. Such interline punctuation could be noted or replaced by a variety of paragraphoi, long marks which trailed between lines of text; these might also mark changes of speakers. Blank lines or various coronides marked the ends of sections. (A separate coronis was used to mark contractions; its early forms looked like an apostrophe between the two ellided words.) Over time, the main punctuation came to be a full stop marked by a single dot at varying heights, a partial stop marked by various forms of commas, and the hypodiastole ⟨⸒⟩ and papyrological hyphen ⟨‿⟩ or ⟨ ͜ ⟩. These served to show whether an ambiguous series of letters should be read as (respectively) a single word or as a pair of words.

Following the advent of printing, most Greek punctuation was gradually standardized with French: the hypodiastole was fully unified with the comma, the comma serves as the decimal point (and in this use is called the "hypodiastole"), the full stop serves as the thousands separator, and guillemets and em-length quotation dashes typically serve to indicate direct speech. The principal difference is the Greek question mark ⟨;⟩, which developed a shape so similar to the Latinate semicolon ⟨;⟩ that Unicode decomposes its separate code point identically. The ano teleia middot serves as the Greek semicolon but is so uncommon that it has often been left off of Greek keyboards.

See also

• Ancient Greek phonology
• Greek alphabet
• Greek braille
• Greek diacritics
• Greek grammar
• Greek language question
• Greek ligatures
• Iota adscript
• Iota subscript
• Katharevousa
• Modern Greek phonology
• Pronunciation of Ancient Greek in teaching
• Romanization of Greek

External links

• Greek Unicode Issues by Nick Nicholas
• The Details of Modern Greek Phonetics and Phonology, by Harry Foundalis