Romanization of Greek is thetransliteration (letter-mapping) ortranscription (sound-mapping) of text from theGreek alphabet into theLatin alphabet.
The conventions forwriting andromanizingAncient Greek andModern Greek differ markedly. The sound of theEnglish letterB (/b/) was written asβ in ancient Greek but is now written as thedigraphμπ, while the modernβ sounds like the English letterV (/v/) instead. TheGreek nameἸωάννης becameJohannes inLatin and thenJohn in English, but in modern Greek has becomeΓιάννης; this might be written asYannis, Jani, Ioannis, Yiannis, or Giannis, but not Giannes or Giannēs as it would be for ancient Greek. The wordΆγιος might variously appear as Hagiοs, Agios, Aghios, or Ayios, or simply betranslated as "Holy" or "Saint" in English forms ofGreek placenames.[1]
Traditional English renderings of Greek names originated fromRoman systems established in antiquity. TheRoman alphabet itself was a form of theCumaean alphabet derived from theEuboean script that valuedΧ as/ks/ andΗ as/h/ and used variant forms ofΛ andΣ that becameL andS.[2] When this script was used to write the classical Greek alphabet, ⟨κ⟩ was replaced with ⟨c⟩, ⟨αι⟩ and ⟨οι⟩ became ⟨æ⟩ and ⟨œ⟩, and ⟨ει⟩ and ⟨ου⟩ were simplified to ⟨i⟩ (more rarely—corresponding to an earlier pronunciation—⟨e⟩) and ⟨u⟩.Aspirated consonants like ⟨θ⟩, ⟨φ⟩, initial-⟨ρ⟩, and ⟨χ⟩ simply wrote out the sound: ⟨th⟩, ⟨ph⟩, ⟨rh⟩, and ⟨ch⟩. BecauseEnglish orthography has changed so much from the originalGreek, modern scholarly transliteration now usually renders ⟨κ⟩ as ⟨k⟩ and the diphthongs ⟨αι, οι, ει, ου⟩ as ⟨ai, oi, ei, ou⟩.[3]
"Greeklish" has also spread withinGreece itself, owing to the rapid spread of digitaltelephony from cultures using theLatin alphabet. Since Greektypefaces andfonts are not always supported or robust,Greek email and chatting has adopted a variety of formats for rendering Greek and Greek shorthand using Latin letters. Examples include "8elo" and "thelw" forθέλω, "3ava" forξανά, and "yuxi" forψυχή.
Owing to the difficulties encountered in transliterating and transcribing both ancient and modern Greek into the Latin alphabet, a number of regulatory bodies have been established. TheHellenic Organization for Standardization (ELOT), in cooperation with theInternational Organization for Standardization (ISO), released a system in 1983 which has since been formally adopted by theUnited Nations, the United Kingdom and United States.
The following tables list several romanization schemes from the Greek alphabet to modern English. Note, however, that the ELOT, UN, and ISO formats for Modern Greek intend themselves as translingual and may be applied in any language using theLatin alphabet.
TheAmerican Library Association andLibrary of Congress romanization scheme employs its "Ancient or Medieval Greek" system for all works and authors up to theFall of Constantinople in 1453,[3] althoughByzantine Greek was pronounced distinctly and some have considered "Modern" Greek to have begun as early as the 12th century.[4]
For treatment ofpolytonic Greek letters—for example,ᾤ—see also thesection on romanizing Greek diacritical marks below.
| Greek | Classical[citation needed] | ALA-LC[3] (2010) | Beta Code[5] |
|---|---|---|---|
| α | a | a | A |
| αι | ae | ai | AI |
| β | b | b | B |
| γ | g | g | G |
| n[n. 1] | n[n. 1] | ||
| δ | d | d | D |
| ε | e | e | E |
| ει | e or i | ei | EI |
| ζ | z | z | Z |
| η | e | ē | H |
| θ | th | th | Q |
| ι | i | i | I |
| κ | c | k | K |
| λ | l | l | L |
| μ | m | m | M |
| ν | n | n | N |
| ξ | x | x | C |
| ο | o | o | O |
| οι | oe | oi | OI |
| ου | u | ou | OU |
| o | |||
| π | p | p | P |
| ρ | rh[n. 2] | rh[n. 2] | R |
| r | r | ||
| σ | s | s | S / S1 |
| ς | S / S2 / J | ||
| τ | t | t | T |
| υ | y | y | U |
| u[n. 3] | u[n. 3] | ||
| υι | ui or yi | ui | UI |
| φ | ph | ph | F |
| χ | ch | ch | X |
| ψ | ps | ps | Y |
| ω | o | ō | W |
ELOT approved in 1982 theELOT 743 standard, revised in 2001,[6] whoseType 2 (Greek:Τύπος 2,romanized: Typos 2)transcription scheme has been adopted by the Greek and Cypriot governments as standard for romanization of names onGreek andCypriot passports. It also comprised aType 1 (Greek:Τύπος 1,romanized: Typos 1)transliteration table, which was extensively modified in the second edition of the standard.
International versions of ELOT 743, with an English language standard document, were approved by the UN (V/19, 1987) and the British and American governments.The ISO approved in 1997 its version,ISO 843, with a differentType 1 transliteration system, which was adopted four years later by ELOT itself, while the U.N. did not update its version. So thetranscriptions of Modern Greek into Latin letters used by ELOT, UN and ISO are essentially equivalent, while there remain minor differences in how they approach reversibletransliteration.
TheAmerican Library Association andLibrary of Congress romanization scheme employs its "Modern Greek" system for all works and authors following theFall of Constantinople in 1453.[3]
In the table below, the special rules for vowel combinations (αι, αυ, ει, ευ, ηυ, οι, ου, ωυ) only apply when these letters function asdigraphs. There are also words where the same letters stand side by side incidentally but represent separate vowels. In these cases each of the two letters is transcribed separately according to the normal rules for single letters. Such cases are marked in Greek orthography by either having anaccent on the first rather than the second vowel letter, or by having adiaeresis(¨ ) over the second letter. For treatment ofaccents anddiaereses—for example,ϊ—also see thesection on romanizing Greek diacritical marks below.
| Greek | IPA | Transcription | Transliteration | Notes | ||||
|---|---|---|---|---|---|---|---|---|
| BGN/PCGN[7] (1962) | ELOT 743 (Type 2 - transcription) (1982; 2001) | UN[8][11] (1987) | ISO[12][7] (1997) | ELOT 743, 2nd ed. (Type 1 - transliteration)[10] (2001) | ALA-LC[3] (2010) | |||
| α | [a] | a | a | a | a | a | a | |
| αι | [e̞] | e | ai | ai | ai | ai | ai | |
| αυ | [av] | av | av | av̱ | au | au | au | before vowels or voiced consonants |
| [af] | af | af | before voiceless consonants and word-finally | |||||
| β | [v] | v | v | v | v | v | v | |
| γ | [ɣ] | g | g | g | g | g | g | |
| [ʝ] | y | before front vowels | ||||||
| γγ | [ŋɡ],[ɲɟ] | ng | ng | ṉg | gg | gg | ng | |
| γκ | [g], [ɟ] | g | gk | gk | gk | gk | gk | word-initially |
| [ŋɡ],[ɲɟ] | ng | ng | word-medially | |||||
| γξ | [ŋks] | nx | nx | ṉx | gx | gx | nx | |
| γχ | [ŋx],[ɲç] | nkh | nch | ṉch | gch | gch | nch | |
| δ | [ð] | dh | d | d | d | d | d | |
| [d] | d | in the combinationνδρ | ||||||
| ε | [e̞] | e | e | e | e | e | e | |
| ει | [i] | i | ei | ei | ei | ei | ei | |
| ευ | [e̞v] | ev | ev | ev̱ | eu | eu | eu | before vowels or voiced consonants |
| [e̞f] | ef | ef | before voiceless consonants and word-finally | |||||
| ζ | [z] | z | z | z | z | z | z | |
| η | [i] | i | i | i̱ | ī | ī / i¯ | ē | |
| ηυ | [iv] | iv | iv | i̱v̱ | īu | īu / i¯u | ēu | before vowels or voiced consonants |
| [if] | if | i̱f̱ | before voiceless consonants and word-finally | |||||
| θ | [θ] | th | th | th | th | th | th | |
| ι | [i] | i | i | i | i | i | i | |
| κ | [k], [c] | k | k | k | k | k | k | |
| λ | [l] | l | l | l | l | l | l | |
| μ | [m] | m | m | m | m | m | m | |
| μπ | [b] | b | b | b | mp | mp | b | word-initially |
| [mb] | mb | mp | mp | mp | word-medially | |||
| ν | [n] | n | n | n | n | n | n | |
| ντ | [d] | d | nt | nt | nt | nt | ḏ / d_ | word-initially |
| [nd] | nd | nt | word-medially and word-finally | |||||
| [nd(z)] | nt | in the combinationντζ | ||||||
| ξ | [ks] | x | x | x | x | x | x | |
| ο | [o̞] | o | o | o | o | o | o | |
| οι | [i] | i | oi | oi | oi | oi | oi | |
| ου | [u] | ou | ou | ou | ou | ou | ou | |
| π | [p] | p | p | p | p | p | p | |
| ρ | [r] | r | r | r | r | r | r | |
| σ / ς | [s] | s | s | s | s | s | s | ς (σίγμα τελικό - final sigma) is used as the final letter in a word. |
| τ | [t] | t | t | t | t | t | t | |
| υ | [i] | i | y | y | y | y | y | |
| υι | [i] | i | yi | yi | yi | yi | ui | |
| φ | [f] | f | f | f | f | f | ph | |
| χ | [x], [ç] | kh | ch | ch | ch | ch | ch | |
| ψ | [ps] | ps | ps | ps | ps | ps | ps | |
| ω | [o̞] | o | o | o̱ | ō | ō / o¯ | ō | |
| ωυ | [oi] | oy | oy | o̱y | ōy | ōy / o¯y | ōu | |
The traditionalpolytonic orthography of Greek usesseveral distinct diacritical marks to render what was originally thepitch accent of Ancient Greek and the presence or absence of word-initial/h/. In 1982,monotonic orthography was officially introduced for modern Greek. The only diacritics that remain are theacute accent (indicating stress) and thediaeresis (indicating that two consecutive vowels should not be combined).
When a Greekdiphthong is accented, the accent mark is placed over thesecond letter of the pair. This means that an accent over thefirst letter of the pair indicates vowels which should be taken (and romanized) separately. Although the second vowel is not marked with a superfluous diaeresis in Greek, the first-edition ELOT 743 and the UN systems place a diaeresis on the Latin vowel for the sake of clarity.[13][8]
| Greek | Ancient | Modern | Name | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Classical | ALA-LC[3] (2010) | Beta Code[5] [n. 1] | ELOT[10] (2001) | UN[8] (1987) | BGN/ PCGN [14] (1996) | ISO[7] (1997) | ALA-LC[3] (2010) | ||
| ́ | / | ´[n. 2] | accent acute accent | ||||||
| ̀ | \ | ` | — | ´ | ` | grave accent | |||
| ῾ | h[n. 3] | ( | h[n. 3] | — | h[n. 3] | h[n. 3] | rough breathing | ||
| ᾿ | ) | ' | — | ' | coronis smooth breathing | ||||
| ˜ ̑ | = | ˆ | — | ´ | ˆ | circumflex | |||
| ¨ | [n. 4] | + | ¨[n. 4] | [n. 4] | diaeresis | ||||
| ͺ | | | ¸ | — | ¸ | iota subscript | ||||
Apart from the diacritical marks native to Greek itself or used to romanize its characters,linguists also regularly markvowel length withmacrons ( ¯ ) markinglong vowels and roundedbreves ( ˘ ) markingshort vowels. Where these are romanized, it is common to mark the long vowels with macrons over the Latin letters and to leave the short vowels unmarked; such macrons should not be confused or conflated with those used by some systems to marketa andomega as distinct fromepsilon,iota, andomicron.
Greece's earlyAttic numerals were based on a small sample of letters (includingheta) arranged in multiples of 5 and 10, likely forming the inspiration for the laterEtruscan andRoman numerals.
This early system was replaced byGreek numerals which employed the entire alphabet, including thenonstandard letters digamma,stigma, or sigma-tau (placed between epsilon and zeta),koppa (placed between pi and rho), andsampi (placed after omega). As revised in 2001, ELOT 743 provides for the uncommon characters to be given (in Greek) as$ for stigma,+ for koppa, and/ for sampi. These symbols are not given lower-case equivalents.[10] When used as numbers, the letters are used in combination with the upperkeraia numeral sign ⟨ʹ⟩ to denote numbers from 1 to 900 and in combination with the lowerkeraia ⟨͵⟩ to denote multiples of 1000. (For a full table of the signs and their values, seeGreek numerals.)
These values are traditionally romanized asRoman numerals, so thatΑλέξανδρος Γ' ο Μακεδών would be translated asAlexander III of Macedon and transliterated asAléxandros III o Makedṓn rather thanAléxandros G' orAléxandros 3.Greek laws and other official documents ofGreece which employ these numerals, however, are to be formally romanized using "decimal"Arabic numerals.[10]
Ancient Greek text did not mark word division withspaces orinterpuncts, instead running the words together (scripta continua). In the Hellenistic period, a variety of symbols arose forpunctuation oreditorial marking; such punctuation (or the lack thereof) are variously romanized, inserted, or ignored in different modern editions.
ModernGreek punctuation generally followsFrench with the notable exception of Greek's use of a separatequestion mark, theerotimatiko, which is shaped like theLatinatesemicolon. Greek punctuation which has been given formal romanizations include:
| Greek | ELOT[10] (2001) | ISO[7] (1997) | Name |
|---|---|---|---|
| ; | ? | ? | Greek question mark (erotimatiko) |
| . | . | . | full stop (teleia) |
| · | ; | ; | Greek semicolon (ano teleia) |
| : | : | : | colon (ano kato teleia) |
| , | , | , | comma (komma) |
| ! | ! | exclamation point (thavmastiko) | |
| ’ | ' | ' | apostrophe (apostrofos) |
| ‿ ͜ | - | - | papyrological hyphen (enotikon) |
There are manyarchaic forms and local variants of theGreek alphabet.Beta, for example, might appear as round Β or pointed
throughout Greece but is also found in the forms
(atGortyn),
and
(Thera),
(Argos),
(Melos),
(Corinth),
(Megara andByzantium), and even
(Cyclades).[15] Well into the modern period, classical and medieval Greek was also set using a wide array ofligatures, symbols combining or abbreviating various sets of letters, such as those included inClaude Garamond's 16th-centurygrecs du roi. For the most part, such variants—asϖ and
forπ,ϛ forστ, andϗ forκαι—are just silently emended to their standard forms and transliterated accordingly. Letters with no equivalent in the classical Greek alphabet such asheta (Ͱ &ͱ), meanwhile, usually take their nearest English equivalent (in this case,h) but are too uncommon to be listed in formal transliteration schemes.
Uncommon Greek letters which have been given formal romanizations include:
| Greek | ISO[7] | ALA-LC[3] | Beta Code[5] | Name |
|---|---|---|---|---|
| Ϝϝ Ͷͷ | w | w | V | digamma |
| Ϙϙ Ϟϟ | — | ḳ | #3 | koppa |
| Ϡϡ Ͳͳ | — | #5 | sampi | |
| Ϻϻ | — | #711 | san | |
| Ϲϲ | s | s | S / S3 | lunatesigma |
| Ϳϳ | j | — | #401 | yot |
The sounds ofModern Greek have diverged from both those of Ancient Greek and their descendant letters in English and other languages. This led to a variety of romanizations for names and placenames in the 19th and 20th century. TheHellenic Organization for Standardization (ELOT) issued its system in cooperation with theInternational Organization for Standardization (ISO) in 1983. This system was adopted (with minor modifications) by theUnited Nations' Fifth Conference on the Standardization of Geographical Names atMontreal in 1987,[8][12] by the United Kingdom'sPermanent Committee on Geographical Names for British Official Use (PCGN) and by the United States'Board on Geographic Names (BGN) in 1996,[14] and by the ISO itself in 1997.[12][16] Romanization of names for official purposes (as with passports and identity cards) were required to use the ELOT system within Greece until 2011, when a legal decision permitted Greeks to use irregular forms[17] (such as "Demetrios" forΔημήτριος) provided that official identification and documents also list the standard forms (as, for example, "Demetrios OR Dimitrios").[18] Other romanization systems still encountered are the BGN/PCGN's earlier 1962 system[12][7] and the system employed by theAmerican Library Association and the United States'Library of Congress.[3]