List of Latin-script digraphs
Letters with diacritics are arranged in alphabetic order according to their base. That is, ⟨å⟩ is alphabetized with ⟨a⟩, not at the end of the alphabet as it would be in Norwegian, Swedish and Danish. Substantially modified letters such as ⟨ſ ⟩ (a variant of ⟨s⟩) and ⟨ɔ⟩ (based on ⟨o⟩) are placed at the end.
⟨’b⟩ (capital ⟨’B⟩) is used in the Bari alphabet for /ɓ/.
⟨’d⟩ (capital ⟨’D⟩) is used in the Bari alphabet for /ɗ/.
⟨aa⟩ is used in the orthographies of Dutch, Finnish and other languages with phonemic long vowels for /aː/. It was formerly used in Danish and Norwegian (and still is in some proper names) for the sound /ɔ/, now spelled ⟨å⟩.
- In Latin orthography, ⟨ae⟩ originally represented the diphthong /ai/, before it was monophthongized in the Vulgar Latin period to /ɛ/; in medieval manuscripts, the digraph was frequently replaced by the ligature ⟨æ⟩.
- In Modern English, Latin loanwords with ⟨ae⟩ are generally pronounced with /iː/ (e.g. Caesar), prompting Noah Webster to shorten this to ⟨e⟩ in his 1806 spelling reform for American English.
- In German orthography, ⟨ae⟩ is a variant of ⟨ä⟩ found in some proper names or in contexts where ⟨ä⟩ is unavailable. In the Dutch alphabet, ⟨ae⟩ is an old spelling variant of the ⟨aa⟩ digraph but now only occurs in names of people or (less often) places and in a few loanwords from Greek.
- In Zhuang, ⟨ae⟩ is used for /a/ (⟨a⟩ is used for /aː/).
⟨ãe⟩ is used in Portuguese orthography for /ɐ̃ĩ̯/.
⟨ai⟩ is used in many languages, typically representing the diphthong /ai/. In English, as a result of the Great Vowel Shift, the vowel of ⟨ai⟩ has shifted from this value to /eɪ/ as in pain and rain; while in French, a different change, monophthongization, has occurred, resulting in the digraph representing /ɛ/. A similar change has also occurred during the development of Greek, resulting in ⟨αι⟩ and the ⟨ε⟩ both having the same sound; originally /ɛ/, later /e/. In German orthography, it represents /aɪ/ as in Kaiser (which derived from Latin caesar). However, most German words use ⟨ei⟩ for /aɪ/. In the Kernowek Standard orthography of Cornish, it represents /eː/, mostly in loanwords from English such as paint.
⟨aí⟩ is used in Irish orthography for /iː/ between a broad and a slender consonant.
⟨aî⟩ is used in French orthography for /ɛː/, as in aînesse /ɛːnɛs/ or maître /mɛːtʁ/.
⟨ái⟩ is used in Irish orthography for /aː/ between a broad and a slender consonant.
⟨ãi⟩ is used in Portuguese orthography for /ɐ̃ĩ̯/. It has, thus, the same value as ⟨ãe⟩, but the latter is much more common.
⟨âm⟩ is used in Portuguese orthography for /ɐ̃/ before a consonant.
⟨an⟩ is used in many languages to write a nasal vowel. In Portuguese orthography it is used for /ɐ̃/ before a consonant, in French it represents /ɑ̃/, and in many West African languages it represents /ã/.
⟨ån⟩ is used in the Walloon language, for the nasal vowel /ɔ̃/.
⟨aŋ⟩ is used in Lakhota for the nasal vowel /ã/
⟨ao⟩ is used in the Irish orthography for /iː/ or /eː/, depending on dialect, between broad consonants. In French orthography, it is found in a few words such as paonne representing /a/. In Malagasy, it represents /o/, and in Piedmontese, /au̯/.
⟨ão⟩ is used in Portuguese orthography for /ɐ̃ũ̯/.
⟨au⟩ in English is a result of various linguistic changes from Middle English, having shifted from */au/ to /ɔː/. In a number of dialects, this has merged with /ɑː/. It occasionally represents the diphthong /aʊ/, as in flautist. Other pronunciations are /æ/ in North American English aunt and laugh, /eɪ/ in gauge, /oʊ/ as in gauche and chauffeur, and /ə/ as in meerschaum and restaurant.
In French orthography, ⟨au⟩ represents /o/ or sometimes /ɔ/. It most frequently appears in the inflectional ending marking plurals of certain kinds of words like cheval ('horse') or canal ('channel'), respectively having a plural in chevaux and canaux. In Icelandic orthography, it represents /œy/.
⟨äu⟩ is used in German orthography for the diphthong /ɔɪ/ in declension of native words with au; elsewhere, /ɔʏ/ is written as ⟨eu⟩. In words where ä|u is separated in two sylables, mostly of Latin origin, ⟨äu⟩ is pronounced as /ɛ.ʊ/, as in Matthäus (one German form for Matthew).
⟨aw⟩ is used in English orthography in ways that parallel English ⟨au⟩, though it appears more often at the end of a word. In Cornish, ⟨aw⟩ represents the diphthong /aʊ/ or /æʊ/. In Welsh orthography, ⟨aw⟩ represents the diphthong /au/.
⟨ay⟩ is used in English orthography in ways that parallel English ⟨ai⟩, though it appears more often at the end of a word.
In French orthography, it is usually used to represent /ɛj/ before a vowel (as in ayant) and /ɛ.i/ before a consonant (as in pays).
⟨bb⟩ is used in Pinyin for /b/ in languages such as Yi, where b stands for /p/. In English, doubling a letter indicates that the previous vowel is short (so bb represents /b/). In ISO romanized Korean, it is used for the fortis sound /p͈/, otherwise spelled ⟨pp⟩; an example is hobbang. In Hadza it is the rare /pʼ/.
⟨bd⟩ is used in English orthography for /d/ in a few words of Greek origin, such as bdellatomy. When not initial, it represents /bd/, as in abdicate.
⟨bh⟩ is used in transcriptions of Indo-Aryan languages for a murmured voiced bilabial plosive (/bʱ/). In Irish orthography, it stands for the phonemes /w/ and /vʲ/, for example mo bhád /mə waːd̪ˠ/ ('my boat'), bheadh /vʲɛx/ ('would be'). In the orthography used in Guinea before 1985, ⟨bh⟩ was used in Pular (a Fula language) for the voiced bilabial implosive /ɓ/, whereas in Xhosa, Zulu, and Shona, ⟨b⟩ represents the implosive and ⟨bh⟩ represents the plosive /b/.
⟨bm⟩ is used in Cornish for an optionally pre-occluded /m/; that is, it is pronounced either /m/ or /mː/ (in any position); /ᵇm/ (before a consonant or finally); or /bm/ (before a vowel); examples are mabm ('mother') or hebma ('this').
⟨cc⟩ is used in Andean Spanish for loanwords from Quechua or Aymara with /q/, as in Ccozcco (modern Qusqu) ('Cuzco'). In many European languages, ⟨cc⟩ before front vowels represents a sequence such as /ks/, e.g. English success, French occire, Spanish accidente (dialectally /ks/ or /kθ/); this is not the case of Italian, where a ⟨cc⟩ before a front vowel represents a geminated /tʃ/, as in lacci /ˈlat.tʃi/. In Hadza it is the glottalized click /ᵑǀˀ/. In Piedmontese and Lombard, ⟨cc⟩ represents the /tʃ/ sound at the end of a word.
⟨cg⟩ is used for the click /ǀχ/ in Naro. It was also used for /dʒ/ in Old English (ecg in Old English sounded like 'edge' in Modern English), and in the Tindall orthography of Khoekhoe for the voiceless dental click /ǀ/.
⟨ch⟩ is used in several languages. In English, it can represent /tʃ/, /k/, /ʃ/, /x/ or /h/. See article.
⟨ck⟩ is used in many Germanic languages in lieu of ⟨kk⟩ or ⟨cc⟩ to indicate either a geminated /kː/, or a /k/ with a preceding (historically) short vowel. The latter is the case with English tack, deck, pick, lock, and buck (compare backer with baker). In German orthography, ⟨ck⟩ indicates that the preceding vowel is short. Prior to the German spelling reform of 1996, it was replaced by ⟨k-k⟩ for syllabification. The new spelling rules allow only syllabification of the ⟨ck⟩ as a whole:
- Old spelling: Säcke: Säk-ke ('sacks')
- New spelling: Säcke: Sä-cke
- Among the modern Germanic languages, ⟨ck⟩ is used mainly in Alsatian, English, German, Luxembourgish, Scots, Swedish, and other West Germanic languages in Austria, Germany and Switzerland. Similarly, ⟨kk⟩ is used for the same purpose in Afrikaans, Danish, Dutch, Icelandic, Norwegian, and other West Germanic languages in the Netherlands and Belgium. Compare the word nickel, which is the same in many of these languages except for the customary ⟨ck⟩ or ⟨kk⟩ spelling. The word is nickel in English and Swedish, Nickel in German, and nikkel in Afrikaans, Danish, Dutch, Icelandic and Norwegian.
- It was also used in the Tindall orthography of Khoekhoe for the voiceless dental click /ǀ/ (equivalent to ⟨cg⟩).
- It is also used in Cornish for /k/ at the end of a syllable after a short vowel; only in loanwords (mostly from English) in the Standard Written Form (SWF), more widely in Kernowek Standard.
⟨cs⟩ is used in the Hungarian alphabet for a voiceless postalveolar affricate, /tʃ/. It is considered a distinct letter, named csé, and is placed between ⟨C⟩ and ⟨D⟩ in alphabetical order. Examples of words with cs include csak ('only'), csésze ('cup'), cső ('pipe').
⟨ct⟩ is used in English orthography for /t/ in a few words of Greek origin, such as ctenoid. When not initial, it represents /kt/, as in act.
⟨cu⟩ is used in the orthographies for languages such as Nahuatl (that is, based on Spanish or Portuguese orthography) for /kʷ/. In Nahuatl, ⟨cu⟩ is used before a vowel, whereas ⟨uc⟩ is used after a vowel.
⟨cz⟩ is used in Polish orthography for /t͡ʂ/ as in cześć (help·info) ('hello'). In Kashubian, ⟨cz⟩ represents /tʃ/. This digraph was once common across Europe (which explains the English spelling of Czech), but has largely been replaced. In French and Catalan, historical ⟨cz⟩ contracted to the ligature ⟨ç⟩, and represents the sound /s/. In Hungarian, it was formerly used for the sound /ts/, which is now written ⟨c⟩.
⟨dc⟩ is used in the orthography of Naro for the click /ᶢǀ/.
⟨dd⟩ is used in English orthography to indicate a /d/ with a preceding (historically) short vowel (e.g. jaded /ˈdʒeɪdəd/ has a "long a" while ladder /ˈlædər/ has a "short a"). In Welsh orthography, ⟨dd⟩ represents a voiced dental fricative /ð/. It is treated as a distinct letter, named èdd, and placed between ⟨D⟩ and ⟨E⟩ in alphabetical order. In the ISO romanization of Korean, it is used for the fortis sound /t͈/, otherwise spelled ⟨tt⟩; examples are ddeokbokki and bindaeddeok. In the Basque alphabet, it represents a voiced palatal plosive /ɟ/, as in onddo, ('mushroom').
⟨dg⟩ is used in English orthography for /dʒ/ in certain contexts, such as with judgement and hedge
⟨dh⟩ is used in the Albanian alphabet, Swahili alphabet, and the orthography of the revived Cornish language for the voiced dental fricative /ð/. The first examples of this digraph are from the Oaths of Strasbourg, the earliest French text, where it denotes the same sound /ð/ developed mainly from intervocalic Latin -t-.
- In early traditional Cornish ⟨ȝ⟩ (yogh), and later ⟨th⟩, were used for this purpose. Edward Lhuyd is credited for introducing the grapheme to Cornish orthography in 1707 in his Archaeologia Britannica. In Irish orthography it represents the voiced velar fricative /ɣ/ or the voiced palatal approximant /j/; at the beginning of a word it shows the lenition of /d̪ˠ/, for example mo dhoras /mˠə ɣoɾˠəsˠ/ ('my door' cf. doras /d̪ˠorˠəsˠ/ 'door'). In the pre-1985 orthography of Guinea, ⟨dh⟩ was used for the voiced alveolar implosive /ɗ/ in Pular, a Fula language. It is currently written ⟨ɗ⟩. In the orthography of Shona it is the opposite: ⟨dh⟩ represents /d/, and ⟨d⟩ /ɗ/. In the transcription of Australian Aboriginal languages, ⟨dh⟩ represents a dental stop, /t̪/.
- In addition, ⟨dh⟩ is used in various romanization systems. In transcriptions of Indo-Aryan languages, for example, it represents the murmured voiced dental plosive /d̪ʱ/ and in the romanization of Arabic, it denotes ⟨ﺫ⟩, which represents /ð/ in Modern Standard Arabic.
⟨dj⟩ is used in the Faroese, French and many French-based orthographies for /dʒ/. In the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara, it represents a postalveolar stop such as /ṯ/ or /ḏ/; this sound is also written ⟨dy⟩, ⟨tj⟩, ⟨ty⟩, or ⟨c⟩.
⟨dl⟩ is used in the Hmong language's Romanized Popular Alphabet for /tˡ/. In the Navajo language orthography, it represents /tɬ/, and in the orthography of Xhosa it represents /ɮ̈/. In Hadza it is ejective /cɬʼ/.
⟨dł⟩ is used in the Tlingit alphabet for /tɬ/ (in Alaska, ⟨dl⟩ is used instead).
⟨dn⟩ is used in Yélî Dnye of Papua New Guinea for nasally released /tn/. In Cornish, it is used for an optionally pre-occluded /n/; that is, it is pronounced either /n/ or /nː/ (in any position); /ᵈn/ (before a consonant or finally); or /dn/ (before a vowel); examples are penn ('head') or pennow ('heads').
⟨dq⟩ is used for the click /ᶢǃ/ in the orthography of Naro.
⟨dy⟩ is used in the Xhosa language orthography for /dʲʱ/. In the Shona alphabet, it represents /dʒɡ/. It is the orthography of Tagalog is used for /dʒ/. In the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara, it represents a postalveolar stop such as /ṯ/ or /ḏ/. This sound is also written ⟨tj⟩, ⟨dj⟩, ⟨ty⟩, ⟨c⟩, or ⟨j⟩.
⟨dz⟩ is used in several languages, often to represent /d͡z/. See article.
⟨dź⟩ is used in the Polish and Sorbian alphabets for /d͡ʑ/, the voiced alveolo-palatal affricate, as in dźwięk /d͡ʑvʲɛŋk/. ⟨Dź⟩ is never written before a vowel (⟨dzi⟩ is used instead, as in dziecko /d͡ʑɛt͡skɔ/ 'child').
⟨ea⟩ is used in many languages. In English orthography, ⟨ea⟩ usually represents the monophthong /i/ as in meat; due to a sound change that happened in Middle English, it also often represents the vowel /ɛ/ as in sweat. Rare pronunciations occur, like /eɪ/ in just break, great, steak, and yea, and /æ/ in the archaic ealdorman. When followed by r, it can represent the standard outcomes of the previously mentioned three vowels in this environment: /ɪər/ as in beard, /ɜr/ as in heard, and /ɛər/ as in bear, respectively; as another exception, /ɑr/ occurs in the words hearken, heart and hearth. It often represents two independent vowels, like /eɪ.ɑː/ (seance), /i.æ/ (reality), /i.eɪ/ (create), and /i.ɨ/ (lineage). Unstressed, it may represent /jə/ (ocean) or /ɨ/ (Eleanor). In the Romanian alphabet, it represents the diphthong /e̯a/ as in beată ('drunk female'). In Irish orthography, ⟨ea⟩ represents /a/ between a slender and a broad consonant. In Old English, it represents the diphthong /æɑ̯/. ⟨Ea⟩ is also the transliteration of the ⟨ᛠ⟩ rune of the Anglo-Frisian Futhorc.
⟨eá⟩ is used in Irish orthography for /aː/ between a slender and a broad consonant.
⟨éa⟩ is used in Irish orthography for /eː/ between a slender and a broad consonant.
⟨ee⟩ represents a long mid vowel in a number of languages. In English orthography, ⟨ee⟩ represents /iː/ as in teen. In both the Dutch and German alphabets, ⟨ee⟩ represents /eː/ (though it is pronounced /eɪ/ in majority of northern Dutch dialects). In Bouyei, ⟨ee⟩ is used for plain /e/, as ⟨e⟩ stands for /ɯ/
⟨ei⟩ usually represents a diphthong. In English orthography, ⟨ei⟩ can represent many sounds, including /eɪ/, as in vein, /i/ as in seize, /aɪ/ as in heist, /ɛ/ as in heifer, /æ/ as in enceinte, and /ɨ/ as in forfeit. See also I before e except after c. In the southern and western Faroese dialects, it represents the diphthong /aɪ/, while in the northern and eastern dialects, it represents the diphthong /ɔɪ/.
In the Welsh alphabet, ⟨ei⟩ represents /əi/. In the Irish and Scottish Gaelic orthographies, it represents /ɛ/ before a slender consonant. In the Dutch alphabet, ⟨ei⟩ represents /ɛi/. In the German alphabet, it represents /aɪ/, as in Einstein. This digraph was taken over from Middle High German writing systems, where it represented /eɪ/. In Modern German, ⟨ei⟩ is predominant in representing /aɪ/, while the equivalent digraph ⟨ai⟩ appears in only a few words. In French orthography, ⟨ei⟩ represents /ɛ/, as in seiche.
⟨eî⟩ is used in French orthography for /ɛː/, as in reître /ʁɛːtʁ/.
⟨éi⟩ is used in Irish orthography for /eː/ between slender consonants.
⟨em⟩ is used in Portuguese orthography for /ɐĩ̯ ~ ẽĩ̯/ at the end of a word and /ẽ/ before a consonant. In French orthography, it can represent /ɑ̃/.
⟨ém⟩ is used in Portuguese orthography for /ɐĩ̯ ~ ẽĩ̯/ at the end of a word.
⟨êm⟩ is used in Portuguese orthography for /ɐĩ̯ ~ ẽĩ̯/ at the end of a word and /ẽ/ before a consonant.
⟨én⟩ is used in Portuguese orthography for /ɐĩ̯ ~ ẽĩ̯/ before a consonant.
⟨ên⟩ is used in Portuguese orthography for /ẽ/ before a consonant.
⟨eo⟩ is used in Irish orthography for /oː/ or occasionally /ɔ/ between a slender and a broad consonant. In the Jyutping romanization of Cantonese, it represents /ɵ/, an allophone of /œː/. In the Revised Romanization of Korean, ⟨eo⟩ represents the open-mid back unrounded vowel /ʌ/, and in Piedmontese it is /ɛu̯/. In English orthography ⟨eo⟩ is a rare digraph without a single pronunciation, representing /ɛ/ in feoff, jeopardy, leopard and the given name Geoffrey, /iː/ in people, /oʊ/ in yeoman and /juː/ in the archaic feodary, while in the originally Gaelic name MacLeod it represents /aʊ/. However, usually it represents two vowels, like /iː.ə/ in leotard and galleon, /iː.oʊ/ in stereo and, /iː.ɒ/ in geodesy, and, uniquely, /uː.iː/ in geoduck.
⟨eu⟩ is found in many languages, most commonly for the diphthong /eu/. Additionally, in English orthography, ⟨eu⟩ represents /juː/ as in neuter (though in yod dropping accents /uː/ may occur). In the German alphabet, it represents /ɔʏ/ as in Deutsch; and in the French, Dutch, Breton, and Piedmontese orthographies, it represents /ø/ as in feu. In Cornish, it represents either long /øː ~ œː/ and short /œ/ or long /eː/ and short /ɛ/. In Yale Cantonese romanization it represents /œː/. In the orthographies of Sundanese and Acehnese, both Austronesian languages, it represents /ɤ/ as in beureum ('red'). In the Revised Romanization of Korean, it represents /ɯ/.
⟨eû⟩ is used in French orthography for /ø/, as in jeûne /ʒøn/.
⟨ew⟩ is used in English orthography for /juː/ as in few and flew. An exception is the pronunciation /oʊ/ in sew, leading to the heteronym sewer,(/ˈsuːər/, 'drain') vs sewer (/ˈsoʊər/, 'one who sews'). In Cornish, it stands for /ɛʊ/.
⟨ey⟩ is used in English orthography for a variety of sounds, including /eɪ/ in they, /iː/ in key, and /aɪ/ in geyser. In the Faroese alphabet, it represents the diphthong /ɛɪ/. In Cornish, it represents the diphthong /ɛɪ/ or /əɪ/.<ref="SWF" />
⟨ff⟩ which may be written as the single unit: ﬀ, is used in English orthography and Cornish for the same sound as single ⟨f⟩, /f/. The doubling is used to indicate that the preceding vowel is (historically) short, or for etymological reasons, in latinisms. Very rarely, ⟨ff⟩ may be found word-initially in English, such as in proper names (e.g. Rose ffrench, Jasper Fforde). In the Welsh alphabet, ⟨ff⟩ represents /f/, while ⟨f⟩ represents /v/. In Welsh, ⟨ff⟩ is considered a distinct letter, and placed between ⟨f⟩ and ⟨g⟩ in alphabetical order. In medieval Breton, vowel nasalisation was represented by a following ⟨ff⟩. This notation was reformed during the 18th century, though proper names retain the former convention, which leads to occasional mispronunciation. For ﬀ as a single unit see: Typographic ligature and Unicode FB00 (U+FB00) in Latin script in Unicode and Unicode equivalence
⟨fh⟩ is used in Irish orthography for the lenition of ⟨f⟩. This happens to be silent, so that ⟨fh⟩ in Irish corresponds to no sound at all. For example, the phrase cá fhad ('how long') is pronounced [kaː ad̪ˠ], where fhad is the lenited form of fad /fɑd/ ('long').
⟨gʻ⟩ is used in the Uzbek orthography to represent /ʁ/ (Cyrillic ⟨ғ⟩). Technically it is not a digraph, since ⟨ʻ⟩ is not a letter of the Uzbek alphabet, but rather a typographic convention for a diacritic. In handwriting the letter is written ⟨ḡ⟩ or ⟨ğ⟩.
⟨ge⟩ is used in French orthography for /ʒ/ before ⟨a o u⟩ as in geôle /ʒol/.
⟨gg⟩ is used in English orthography for /ɡ/ before ⟨i⟩ and ⟨e⟩. It is also used in Pinyin for /ɡ/ in languages such as Yi. In the orthography of Central Alaskan Yup'ik, it represents /x/. In Greenlandic orthography, it represents /çː/. In the ISO romanization of Korean, it is used for the fortis sound /k͈/, otherwise spelled ⟨kk⟩ (e.g. ggakdugi). In Hadza it is ejective /kxʼ/. In Italian, ⟨gg⟩ before a front vowel represents a geminated /dʒ/, as in legge /ˈled.dʒe/. In Piedmontese and Lombard, ⟨gg⟩ is an etymological spelling representing an /tʃ/ at the end of a word which is the unvoicing of an ancient /dʒ/.
⟨gh⟩ is used in several languages. In English, it can be silent or represent /ɡ/ or /f/. See article.
⟨gj⟩ is used in the Albanian alphabet for the voiced palatal plosive /ɟ/, though for Gheg speakers it represents /dʒ/. In the Arbëresh dialect, it represents the voiced velar plosive /ɡʲ/. In the Norwegian and Swedish alphabets, ⟨gj⟩ represents /j/ in words like gjorde ('did'). In Faroese, it represents /dʒ/. It is also used in the Romanization of Macedonian as a Latin equivalent of Cyrillic ⟨Ѓ⟩.
⟨gl⟩ is used in the Italian alphabet for /ʎ/ before ⟨i⟩. Elsewhere /ʎ/ is represented by the trigraph ⟨gli⟩.
⟨gm⟩ is used in English orthography for /m/ in a few words of Greek origin, such as phlegm and paradigm. Between vowels, it simply represents /ɡm/, as in paradigmatic.
⟨gn⟩ is used in the Latin orthography, where it represented /ŋn/ in the classical period. Latin velar-coronal sequences like this (and also ⟨cl cr ct gd gl gr x⟩) underwent a palatal mutation to varying degrees in most Italo-Western Romance languages. For most languages that preserve the ⟨gn⟩ spelling (such as Italian and French), it represents a palatal nasal /ɲ/. This was not the case in Dalmatian and the Eastern Romance languages where a different mutation changed the velar component to a labial consonant as well as the spelling to ⟨mn⟩.
- In English orthography, ⟨gn⟩ represents /n/ initially and finally (i.e. gnome, gnu, benign, sign). When it appears between two syllables, it represents /ɡn/ (e.g. signal). In the Norwegian and Swedish alphabets, ⟨gn⟩ represents /ŋn/ in monosyllabic words like agn, and between two syllables, tegne. Initially, it represents /ɡn/, e.g. Swedish gnista /ˈɡnɪsta/.
⟨go⟩ is used in the Piedmontese alphabet for /ɡw/.
⟨gr⟩ is used in the orthography for Xhosa for /ɣ̈/.
⟨gu⟩ is used in the English, Spanish, French, Portuguese and Catalan orthographies for /ɡ/ before front vowels ⟨i e⟩ (⟨i e y⟩ in English and French) where a "soft g" pronunciation (English /dʒ/; Spanish /x/; French, Portuguese and Catalan /ʒ/) would otherwise occur. In English, it can also be used to represent /ɡw/. In the Ossete Latin alphabet, it is used for /ɡʷ/.
⟨gw⟩ is used in various languages for /ɡʷ/, and in the orthography for Dene Suline it represents /kʷ/.
⟨gy⟩ is used in the Hungarian alphabet for a voiced palatal plosive /ɟ/. In Hungarian, the letter's name is gyé. It is considered a single letter, and acronyms keep the digraph intact. The letter appears frequently in Hungarian words, such as the word for "Hungarian" itself: magyar. In the old orthography of Bouyei, it was used for /tɕ/.
⟨hh⟩ is used in the Xhosa language to write the murmured glottal fricative /ɦ̤/, though this is often written h. In the Iraqw language, hh is the voiceless epiglottal fricative /ʜ/, and in Chipewyan it is a velar/uvular /χ/. In Esperanto, it is an official surrogate of ĥ.
⟨hu⟩ is used primarily in the Classical Nahuatl language, in which it represents the /w/ sound before a vowel; for example, Wikipedia in Nahuatl is written Huiquipedia. After a vowel, ⟨uh⟩ is used. In the Ossete Latin alphabet, hu was used for /ʁʷ/, similar to French roi. The sequence hu is also found in Spanish words such as huevo or hueso; however, in Spanish this is not a digraph but a simple sequence of silent h and the vowel u.
⟨ie⟩ is found in English, where it usually represents the /aɪ/ sound as in pries and allied or the /iː/ sound as in priest and rallied. Followed by an r, these vowels follow the standard changes to /aɪə/ and /ɪə/, as in brier and bier. Unique pronunciations are /ɪ/ in sieve, /ɛ/ in friend and /eɪ/ in lingerie. Unstressed it can represent /jə/, as in spaniel and conscience, or /ɨ/ or /ə/ as in mischief and hurriedly. It also can represent many vowel combinations, including /aɪə/ in diet and client, /aɪɛ/ in diester and quiescent, /iːə/ in alien and skier, /iːɛ/ in oriental and hygienic, and /iːʔiː/ in British medieval.
- In Dutch, the ⟨ie⟩ represents /i/. In German, it may represent the lengthened vowel /iː/ as in Liebe (love) as well as the vowel combination /iə/ as in Belgien (Belgium). In Latvian and Lithuanian, the ⟨ie⟩ is considered two letters for all purposes and represents /iæ̯/, commonly (although less precisely) transcribed as /i̯e/. In Maltese, ⟨ie⟩ is a distinct letter and represents a long close front unrounded vowel, /iː/) or /iɛ/. In Pinyin it is used to write the vowel /e/ in languages such as Yi, where e stands for /ɛ/.
⟨ig⟩ is used in Catalan for /t͡ʃ/ in the coda.
⟨ih⟩, in the practical orthography of the Taa language, represents the breathy or murmured vowel /i̤/. It is also used in Tongyong Pinyin and Wade-Giles transcription for the fricative vowels of Mandarin Chinese, which are spelled i in Hanyu Pinyin.
⟨ii⟩ is used in many languages with phonemic long vowels for /iː/.
⟨il⟩ is used in French for /j/, historically /ʎ/, as in ail /aj/ "garlic".
⟨im⟩ is used in Portuguese orthography for /ĩ/.
⟨ím⟩ is used in Portuguese orthography for /ĩ/ before a consonant.
⟨ín⟩ is used in Portuguese orthography for /ĩ/ before a consonant.
⟨în⟩ is used in French to write a vowel sound /ɛ̃/ that was once followed by a historical s, as in vous vîntes /vu vɛ̃t/ "you came".
⟨iŋ⟩ is used in Lakhota for the nasal vowel /ĩ/.
⟨io⟩ is used in Irish for /ɪ/, /ʊ/, and /iː/ between a slender and a broad consonant.
⟨ío⟩ is used in Irish for /iː/ between a slender and a broad consonant.
⟨iú⟩ is used in Irish for /uː/ between a slender and a broad consonant.
⟨ix⟩ is used in Catalan for /ʃ/ after a vowel.
⟨jh⟩ is used in Walloon to write a sound that is variously /h/ or /ʒ/, depending on the dialect. In Tongyong pinyin, it represents /tʂ/, written zh in standard pinyin. Jh is also the standard transliteration for the Devanāgarī letter झ /dʒʱ/. In the official Esperanto orthography, it is a surrogate of ĵ.
⟨kh⟩, in transcriptions of Indo-Aryan languages, represents the aspirated voiceless velar plosive (/kʰ/). For scores of other languages, it represents the voiceless velar fricative /x/, for example in transcriptions of the letter ḥāʼ (خ) in standard Arabic, standard Persian, and Urdu, Cyrillic Х, х (kha), Spanish j, etc. As the transcription of the letter ḥet (ח) in Sephardic Hebrew, it represents the voiceless pharyngeal fricative /ħ/. It is also used to transcribe the Hebrew letter kaf (כ) in instances when the letter is lenited. When transliterating Russian, Ukrainian, Belarusian and Bulgarian, all written only in the Cyrillic alphabet, the diagraph is equivalent to the Cyrillic letter Х.
- In Canadian Tlingit it represents /qʰ/, which in Alaska is written k. In the Ossete Latin alphabet, it was used for /kʼ/.
⟨kk⟩ is used in Icelandic and Faroese for the pre-aspirated sound /ʰk/, in romanized Korean for the fortis sound /k͈/, and in Haida (Bringhurst orthography) for ejective /kʼ/,in Finnish for a geminate constant /kː/.
⟨kl⟩ is used in the Zulu language to write a sound variously realized as /kʟ̥ʼ/ or /kxʼ/.
⟨kr⟩ is used in the Xhosa language for /kxʼ/.
⟨kv⟩ is used for /kʷʰ/ in some dialects of Zhuang.
⟨kw⟩ is used in various languages for the labialized velar consonant /kʷ/, and in Dene Suline (Chipewyan) for /kʷʰ/. Used informally in English for phonemic spelling of qu, as in kwik (from quick), ultimately from Proto-Indo-European */kʷ/.
⟨ky⟩ is used in Tibetan Pinyin for /tʃʰ/.
⟨kz⟩ is used for /ɡz/ in Esperanto, though some speakers pronounce it /kz/.
⟨lh⟩, in Occitan, Gallo, and Portuguese, represents a palatal lateral approximant /ʎ/. In many Indigenous languages of the Americas it represents a voiceless alveolar lateral fricative /ɬ/. In the transcription of Australian Aboriginal languages it represents a dental lateral, /l̪/. In the Gwoyeu Romatzyh romanization of Mandarin Chinese, initial ⟨lh⟩ indicates an even tone on a syllable beginning in /l/, which is otherwise spelled ⟨l⟩.
⟨lj⟩ is a letter in some Slavic languages, such as the Latin orthographies of Serbo-Croatian, where it represents a palatal lateral approximant /ʎ/. For example, the word ljiljan is pronounced /ʎiʎan/. Ljudevit Gaj first used the digraph ⟨lj⟩ in 1830; he devised it by analogy with a Cyrillic digraph, which developed into the ligature љ.
- The sound /ʎ/ is written ⟨gl⟩ in Italian, in Castilian Spanish and Catalan as ⟨ll⟩, in Portuguese as ⟨lh⟩, in some Hungarian dialects as ⟨lly⟩, and in Latvian as ⟨ļ⟩. In Czech and Slovak, it is often transcribed as ⟨ľ⟩; it is used more frequently in the latter language. There are dedicated Unicode glyphs, ǉ, ǈ, and Ǉ.
⟨ḷḷ⟩ is used in Asturian for a sound that was historically /ʎ/ but which is now an affricate, /t͡s], [t͡ʃ], [d͡ʒ/.
⟨lw⟩ is used for /lʷ/ in Arrernte.
⟨mb⟩, in many African languages, represents /mb/ or /ᵐb/. It is used in Irish to indicate the eclipsis of b and represents /mˠ/; for example ár mbád /ɑːɾˠ mˠɑːdˠ̪/ "our boat" (cf. /bˠɑːd̪ˠ/ "boat"). The Irish digraph is capitalized mB, for example i mBaile Átha Cliath "in Dublin". In English, mb represents /m/ when final, as in lamb. In Standard Zhuang and in Bouyei, ⟨mb⟩ is used for /ɓ/.
⟨mg⟩ is used in Pinyin for /ŋɡ/ in languages such as Yi, where the more common diacritic ⟨ng⟩ is restricted to /ŋ/. It is used in Yélî Dnye of Papua New Guinea for doubly articulated and prenasalized /ŋ͡mk͡p/.
⟨mh⟩, in Irish, stands for the lenition of ⟨m⟩ and represents /v/ or /w/; for example mo mháthair /mə ˈwɑːhəɾʲ/ or /mˠə ˈvˠɑːhəɾʲ/ "my mother" (cf. máthair /ˈmˠɑːhəɾʲ/ "mother"). In Welsh it stands for the nasal mutation of ⟨p⟩ and represents /m̥/; for example fy mhen /ə m̥ɛn/ "my head" (cf. pen /pɛn/ "head"). In both languages it is considered a sequence of the two letters ⟨m⟩ and ⟨h⟩ for purposes of alphabetization. It also occurs in Shona. In the Gwoyeu Romatzyh romanization of Mandarin Chinese, initial mh- indicates an even tone on a syllable beginning in /m/, which is otherwise spelled m-.
⟨mm⟩ is used in Haida (Bringhurst orthography) for glottalized /ˀm/. It is used in Cornish for an optionally pre-occluded /m/; that is, it is pronounced either /m/ or /mː/ (in any position); /ᵇm/ (before a consonant or finally); or /bm/ (before a vowel); examples are mabm ('mother') or hebma ('this').
⟨mn⟩ is used in English to write the word-initial sound /n/ in a few words of Greek origin, such as mnemonic. When final, it represents /m/, as in damn, and between vowels it represents /mn/, as in damnation. In French it represents /n/, as in automne and condamner.
⟨mw⟩ is used for /mʷ/ in Arrernte.
⟨nd⟩, in many African languages, represents /nd/ or /ⁿd/, and capitalized ⟨Nd⟩. It is used in Irish for the eclipsis of ⟨d⟩, and represents /n/, for example in ár ndoras /ɑːɾˠ ˈnˠɔɾˠəsˠ/ "our door" (cf. doras /ˈd̪ˠɔɾˠəsˠ/ "door"). In this function it is capitalized ⟨nD⟩, e.g. i nDoire "in Derry". In Standard Zhuang and in Bouyei, ⟨nd⟩ is used for /ɗ/.
⟨nf⟩, equivalent to ⟨mf⟩ for /mf/ or /ᵐf/.
⟨ng⟩, in English and several other European and derived orthographies (for example Vietnamese), generally represents the velar nasal /ŋ/. It is considered a single letter in many Austronesian languages (Māori, Tagalog, Tongan, Kiribatian, Tuvaluan, Indonesian), the Welsh language, and Rheinische Dokumenta, for velar nasal /ŋ/; and in some African languages (Lingala, Bambara, Wolof) for prenasalized /ɡ/ (/ⁿɡ/).
- The Finnish language uses the digraph 'ng' to denote the phonemically long velar nasal /ŋː/ in contrast to 'nk' /ŋk/, which is its "strong" form under consonant gradation, a type of lenition. Weakening /k/ produces an archiphonemic "velar fricative", which, as a velar fricative does not exist in Standard Finnish, is assimilated to the preceding /ŋ/, producing /ŋː/. (No /ɡ/ is involved at any point, despite the spelling 'ng'.) The digraph 'ng' is not an independent letter, but it is an exception to the phonemic principle, one of the few in standard Finnish.
- In Irish ng is used word-initially as the eclipsis of g and represents /ŋ/, e.g. ár ngalar /ɑːɾˠ ˈŋɑɫəɾˠ/ "our illness" (cf. /ˈɡɑɫəɾˠ/). In this function it is capitalized nG, e.g. i nGaillimh "in Galway".
- In Tagalog and other Philippine languages, ng represented the prenasalized sequence /ŋɡ/ during the Spanish era. The velar nasal, /ŋ/, was written in a variety of ways, namely "n͠g", "ñg", "gñ" (as in Sagñay), and—after a vowel—at times "g̃". During the standardization of Tagalog in the early part of the 20th century, ng became used for the velar nasal /ŋ/, while prenasalized /ŋɡ/ came to be written ngg. Furthermore, ng is also used for a common genitive particle pronounced /naŋ/, to differentiate it from an adverbial particle nang.
⟨ńg⟩ is used in Central Alaskan Yup'ik to write the voiceless nasal sound /ŋ̊/.
⟨ñg⟩, or more precisely ⟨n͠g⟩, was a digraph in several Spanish-derived orthographies of the Pacific, such as that of Tagalog and Chamorro, where it represented the sound /ŋ/, as opposed to ng, which originally represented /ŋɡ/. An example is Chamorro agan͠gñáijon (modern agangñaihon) "to declare". Besides ñg, variants of n͠g include gñ (as in Sagñay), ng̃, and a g̃, that is preceded by a vowel (but not a consonant). It has since been replaced by the trigraph ⟨ngg⟩ or ⟨ng⟩ (see above).
⟨ng’⟩ is used for /ŋ/ in Swahili and languages with Swahili-based orthographies. Since ⟨’⟩ is not a letter in Swahili, ⟨ng’⟩ is technically a digraph, not a trigraph.
⟨nh⟩ is used in several languages. See article.
⟨nj⟩ is a letter present in the Latin orthographies of Albanian, Serbo-Croatian. Ljudevit Gaj, a Croat, first used this digraph in 1830. In all of these languages, it represents the palatal nasal /ɲ/. For example, the Croatian and Serbian word konj (horse) is pronounced /koɲ/. The digraph was created in the 19th century by analogy with a digraph of Cyrillic, which developed into the ligature ⟨Њ⟩. There are dedicated glyphs in Unicode, Ǌ, ǋ, ǌ.
- In Faroese, it generally represents /ɲ/, although in some words it represent /nj/, like in banjo. It is also used in some languages of Africa and Oceania where it represents a prenazalized voiced postalveolar affricate or fricative, /ⁿdʒ/ or /ⁿʒ/. In Malagasy, it represents /ⁿdz/.
- Other letters and digraphs of the Latin alphabet used for spelling this sound are ⟨ń⟩ (in Polish), ⟨ň⟩ (in Czech and Slovakian), ⟨ñ⟩ (in Spanish), ⟨nh⟩ (in Portuguese and Occitan), ⟨gn⟩ (in Italian and French), and ⟨ny⟩ (in Hungarian, among others).
⟨nk⟩ is used in the orthography of many Bantu languages like Lingala, Tshiluba, and Kikongo, for /ŋk/ or /ᵑk/. In the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara, it distinguishes a prenasalized velar stop, /ŋ͡k ~ ŋ͡ɡ/, from the nasal /ŋ/.
⟨nn⟩ is used in Irish orthography for the Old Irish "fortis sonorants" /Nˠ/ ("broad", i.e. non-palatalized or velarized) and /Nʲ/ ("slender", i.e. palatalized) in non-initial position. In modern Irish, the "broad" sound is /n̪ˠ/, while the slender sound can be any of /nʲ/, /n̠ʲ/, or /ɲ/, depending on dialect and position in the word. In Spanish historical ⟨nn⟩ has contracted to the ligature ⟨ñ⟩ and represents the sound /ɲ/. In the Gwoyeu Romatzyh romanization of Mandarin Chinese, final -nn indicates a falling tone on a syllable ending in /n/, which is otherwise spelled -n. It is used in Haida (Bringhurst orthography) for glottalized /ˀn/. In Piedmontese, it is /ŋn/ in the middle of a word, and /n/ at the end. In Cornish, it is used for an optionally pre-occluded /n/; that is, it is pronounced either /n/ or /nː/ (in any position); /ᵈn/ (before a consonant or finally); or /dn/ (before a vowel); examples are penn ('head') or pennow ('heads').
⟨nq⟩ is used in various alphabets. In the Romanized Popular Alphabet used to write Hmong, it represents the sound /ɴɢ/. In Xhosa and Zulu it represents the click /ᵑǃ/. In the Gwoyeu Romatzyh romanization of Mandarin Chinese, final -nq indicates a falling tone on a syllable ending in /ŋ/, which is otherwise spelled -ng.
⟨nv⟩, equivalent to ⟨mv⟩ for /mv/ or /ᵐv/.
⟨ny⟩ is used in several languages for /ɲ/. See article.
⟨n-⟩ is used for medial /ŋ/ in Piedmontese.
⟨o′⟩, in the practical orthography of the Taa language, represents the glottalized or creaky vowel /o̰/. It is also used for /o/ and /ø/ in Romanized Uzbek, with the preferred typographical form being ⟨oʻ⟩ (Cyrillic ⟨ў⟩). Technically it is not a digraph in Uzbek, since ⟨ʻ⟩ is not a letter of the Uzbek alphabet, but rather a typographic convention for a diacritic. In handwriting the letter is written ⟨ō⟩ or ⟨ŏ⟩.
⟨oa⟩ is used in English, where it commonly represents the /oʊ/ sound as in road, coal, boast, coaxing, etc. In Middle English, where the digraph originated, it represented /ɔː/, a pronunciation retained in the word broad and derivatives, and when the digraph is followed by an "r", as in soar and bezoar. The letters also represent two vowels, as in koala /oʊ.ɑː/, boas /oʊ.ə/, coaxial /oʊ.æ/, oasis /oʊ.eɪ/, and doable /uː.ə/. In Malagasy, it is occasionally used for /o/.
⟨oe⟩ is found in many languages. In English, it represents the /oʊ/ sound as in hoe and sometimes the /uː/ sound as in shoe. Afrikaans and Dutch oe is /u/, as in doen. Often ligatured to ⟨œ⟩ in French, it stands for the vowels /œ/ (as in œil /œj/) and /e/ (as in œsophage /ezɔfaʒ ~ øzɔfaʒ/). It is an alternative way to write ⟨ö⟩ in German when this character is unavailable. In Cantonese Pinyin it represents the vowel /ɵ] ~ [œː/, and in Zhuang it is used for /o/ (⟨o⟩ is used for /oː/). In Piedmontese, it is /wɛ/. In the Kernewek Kemmyn orthography of Cornish, it is used for a phoneme which is [oː] long, [oˑ] mid-length, and [ɤ] short.
⟨oê⟩ is used in French to write the vowel sound /wa/ in a few words before what had historically been an s, mostly in words derived from poêle /pwal/ "stove". The diacriticless variant, ⟨oe⟩, rarely represents this sound except in words related to moelle /mwal/ (rarely spelt moëlle).
⟨oi⟩ is used in various languages. In English, oi represents the /oɪ̯/ sound as in coin and join. In French, it represents /wa/, which was historically – and still is in some cases – written "oy." In Irish it is used for /ɛ/, /ɔ/, /ɪ/, /əi̯/, /iː/, /oː/ between a broad and a slender consonant. In Piedmontese, it is /ui̯/.
⟨oí⟩ is used in Irish for /iː/ between a broad and a slender consonant.
⟨oî⟩ is used in French to write /wa/ before what had historically been an s, as in boîtier or cloître.
⟨ói⟩ is used in Irish for /oː/ between a broad and a slender consonant.
⟨òi⟩ is used in Piedmontese for /oi̯/.
⟨om⟩ is used in Portuguese orthography for /õ/.
⟨ôm⟩ is used in Portuguese orthography for /õ/ before a consonant.
⟨on⟩ is used in Portuguese orthography for /õ/ before a consonant, and in French to write /ɔ̃/.
⟨ôn⟩ is used in Portuguese orthography for /õ/ before a consonant.
⟨oo⟩ is used in many languages. In English, oo commonly represents two sounds: /uː/ as in "moon" and "food", and /ʊ/ as in "wood" and "foot". Historically, both derive from the sound /oː/, which is also the digraph’s pronunciation in most other languages. In German and Dutch, the digraph represents /oː/. In Cornish, it represents either /oː/ or /uː/.
⟨ou⟩ is used in English for the diphthong /aʊ/, as in out /aʊt/. This spelling is generally used before consonants, with ⟨ow⟩ being used instead before vowels and at the ends of words. Occasionally ou may also represent other vowels – /ʌ/ as in trouble, /oʊ/ as in soul, /ʊ/ as in would, or /uː/ as in group. The ou in out originally represented /uː/, as in French, but its pronunciation has changed as part of the Great Vowel Shift.
In Dutch ⟨ou⟩ represents /ʌu/ in the Netherlands or /oʊ/ in Flanders. In Cornish, it represents [uː], [u], or [ʊ]. In French, it represents the vowel /u/, as in vous /vu/ "you", or the approximant consonant /w/, as in oui /wi/ "yes".
⟨oû⟩ is used in French to write the vowel sound /u/ before what had historically been an s, as in soûl /sul/ "drunk".
⟨ow⟩, in English, usually represents the /aʊ/ sound as in coward, sundowner, and now or the /oʊ/ sound, as in froward, landowner, and know. An exceptional pronunciation is /ɒ/ in knowledge and rowlock. There are many English heteronyms distinguished only by the pronunciation of this digraph, like: bow (front of ship or weapon), bower (a dwelling or string player), lower (to frown or drop), mow (to grimace or cut), row (a dispute or line-up), shower (rain or presenter), sow (a pig or to seed), tower (a building or towboat). In Cornish, this represents the diphthong /ɔʊ/ or /oʊ/; before vowels, it can also represent /uː/.
⟨oy⟩ is found in many languages. In English and Faroese, oy represents the diphthong /ɔɪ/. Examples in English include toy and annoy. In Cornish, it represents the diphthong /oɪ/~/ɔɪ/; in the words oy ('egg') and moy ('much'), it can also be pronounced /uɪ/~/ʊɪ/.
⟨oŷ⟩ is an obsolete digraph once used in French.
⟨øy⟩ is used in Norwegian for /øʏ/.
⟨pf⟩ in German represents a labial affricate /pf/. It can be initial (Pferd, 'horse'), medial (Apfel, 'apple'), or final (Knopf, 'button').
Where it appears in English, usually in names or words recently derived from German, it is ordinarily simplified to ⟨f⟩.
⟨ph⟩, in the English Language and many other languages, represents /f/. Ph in English generally occurs in words derived from Greek, due to Latin transcription of Ancient Greek phi (Φ φ) as ⟨ph⟩, which originally represented /pʰ/ (an aspirated voiceless bilabial plosive). In some non-standard spellings of English, like leet, ph may be used as a replacement of all occurrences of f. Exceptionally, ⟨ph⟩ represents /v/ in the name Stephen and some speakers' pronunciation of the word nephew.
The French and German languages and the international auxiliary languages Interlingua and Occidental also use the digraph for Greek loanwords. In German, ph can be replaced by f; the replacement is allowed in certain cases according to the German spelling reform of 1996. In most Romance (such as Spanish) and Germanic languages, f is used in place of ph. Languages written in a Cyrillic script, such as Russian, Ukrainian, and Bulgarian, regularly use Ф ф – similar to the Greek Φ φ – where the Romance and Germanic languages use ph or f. In Welsh, ph represents /f/ in native words, but only word-initially as the result of an initial consonant mutation of a word beginning with p. Irish uses f for words of Greek origin, while ph represents the lenited form of p, resulting in the sound /f/ as well. In Vietnamese, ph is used exclusively; the letter f is not used. In Old High German, ph stands for the affricate /pf/. In the romanizations of Indo-Aryan languages and of Thai, ph represents the aspirated sound /pʰ/. In the Ossete Latin alphabet, it was used for /pʼ/.
⟨pm⟩ is used for /ᵖm/ in Arrernte.
⟨pn⟩ is used in English for an initial sound /n/ in words of Greek origin such as pneumatic. When not initial, it represents the sequence /pn/, as in apnea.
⟨pp⟩ is used in romanized Korean for the fortis sound /p͈/.
⟨ps⟩ is used in English for an initial sound /s/ in words of Greek origin such as psyche. When not initial, it represents the sequence /ps/, as in ellipse. It is also used in the Shona language to write a whistled sibilant cluster /ps͎/.
⟨pt⟩ is used in several languages for /t/ in words of Greek origin, where it was /pt/. An example in English is pterosaur /ˈtɛrəsɔr/, and an exception is ptarmigan /ˈtɑrmɨɡən/, which is Gaelic, not Greek. When not initial, pt represents the sequence /pt/, as in apt.
⟨pw⟩ is used for /pʷ/ in Arrernte.
⟨qo⟩ is used in Piedmontese for /kw/.
⟨qq⟩ is used in Haida (Bringhurst orthography) for ejective /qʼ/. In Hadza it is the glottalized click /ᵑǃˀ/.
⟨qu⟩ is used in Catalan, French, Galician, Occitan, Portuguese and Spanish orthographies for /k/ before the vowel letters e, i, where the letter c represents the sound /θ/ (Castilian Spanish and most of Galicia) or /s/ (Catalan, French, Latin American Spanish, Occitan and Portuguese). This dates to Latin qu, and ultimately the Proto-Indo-European labialized velar consonant */kʷ/; in English this sound instead became written primarily as wh, due to Grimm's law changing kʷ > xʷ (written hw), and Middle English spelling change switching hw to wh. In English, it represents /k/ in words derived from those languages (e.g., quiche), and /kw/ in other words, including borrowings from Latin (e.g., quantity). In the Ossete Latin alphabet, it was used for /qʷ/. In Vietnamese it was used to represent the /kw/ or /w/ sound. In Cornish, it represents the /kw/ sound.
⟨qv⟩ is used for glottalized /ˀw/ in Bouyei.
⟨qw⟩ is used in some languages for the sound /qʷ/. In Mi'kmaq it is used for /xʷ/. In the Kernowek Standard and Standard Written Form orthographies for Revived Cornish, and in William Jordan's 1611 Creation of the World, it is used for /kw/.
⟨qy⟩ is used for glottalized /ˀj/ in Bouyei.
⟨rh⟩ is found in English language with words from the Greek language and transliterated through the Latin language. Examples include "rhapsody", "rhetoric" and "rhythm". These were pronounced in Ancient Greek with a voiceless "r" sound, /r̥/, as in Old English ⟨hr⟩. The digraph may also be found within words, but always at the start of a word component, e.g., "polyrhythmic". German, French, and the auxiliary language Interlingua use rh in the same way. ⟨Rh⟩ is also found in the Welsh language where it represents a voiceless alveolar trill (r̥), that is a voiceless "r" sound. It can be found anywhere; the most common occurrence in the English language from Welsh is in the slightly respelled given name "Rhonda". In Wade-Giles transliteration, ⟨rh⟩ is used for the syllable-final rhotic of Mandarin Chinese. In the Gwoyeu Romatzyh romanization of Mandarin Chinese, initial rh- indicates an even tone on a syllable beginning in /ʐ/, which is otherwise spelled r-. In Purépecha, it is a retroflex flap, /ɽ/.
⟨rl⟩ is used in the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara for a retroflex lateral, written /ɭ/ in the IPA. In the Greenlandic language, it represents /ɬː/ as the result of an assimilation of a consonant cluster with a uvular consonant as the first component.
⟨rm⟩ is used in Inuktitut for /ɴm/.
⟨rn⟩ represents the retroflex nasal /ɳ/ in Warlpiri, Arrernte, and Pitjantjatjara (see transcription of Australian Aboriginal languages). In the Greenlandic language, it represents /ɴ/. In Inuktitut, it represents /ɴn/.
⟨rp⟩ is used in the Greenlandic language for /pː/ as the result of an assimilation of a consonant cluster with a uvular consonant as the first component.
⟨rr⟩ is used in English language for ⟨r⟩, depending on etymology. It normally appears in words of Latin or Romance origin, and "rrh" in words of ancient Greek origin. It is quite a common digraph, found in words as diverse as arrest, carry, and sorry. Some words with "rr" are relatively recent loanwords from other languages; examples include burro from Spanish. It is often used in impromptu pronunciation guides to denote either an alveolar tap or an alveolar trill. It is a letter in the Albanian alphabet.
In several European languages, such as Catalan, Spanish, Portuguese or Albanian, "rr" represents the alveolar trill /r/ (or the voiced uvular fricative /ʁ/ in Portuguese) and contrasts with the single "r", which represents the alveolar tap /ɾ/ (in Catalan and Spanish a single "r" also represents the alveolar trill at the beginning of words or syllables). In Italian or Finnish, "rr" is furthermore a geminate (long) consonant /rː/. In Central Alaskan Yup'ik it is used for /χ/. In Cornish, it can represent either /rː/, /ɾʰ/, or /ɹ/.
⟨rs⟩ was equivalent to ⟨rz⟩ and stood for /r̝/ (modern ř) in medieval Czech. In the Greenlandic language, it represents /sː/ as the result of an assimilation of a consonant cluster with a uvular consonant as the first component.
⟨rw⟩ is used for /ɻʷ/ in Arrernte.
⟨rz⟩ is used in Polish and Kashubian for a voiced retroflex fricative ʐ, similar to English "zh" as in Zhivago. Examples from Polish are marzec (help·info) "March" and rzeka (help·info) "river". ⟨Rz⟩ represents the same sound as ⟨ż⟩, the only difference being that ⟨ż⟩ evolved from a *g while ⟨rz⟩ is descended from a palatalized ar ( *rʲ ). ⟨Rz⟩ usually corresponds to Czech ⟨ř⟩, though the pronunciations are different. When preceded by a voiceless consonant (ch, k, p, t) or end of a word, ⟨rz⟩ devoices to ⟨sz⟩, as in przed (help·info) "before", pronounced [ˈpʂɛt].
⟨sc⟩ is used in Italian for /ʃː/ before the front vowel letters e, i. It is used for /s/ in Catalan, French, English, Occitan and Portuguese (e.g. French/English reminiscence, Portuguese reminiscência, Catalan reminiscència, Occitan reminiscéncia).
⟨sg⟩ is used in Piedmontese for /ʒ/.
⟨sh⟩ is used in several languages. In English, it represents /ʃ/. See article. See also ſh below, which has the capitalized forms SH and ŞH.
⟨si⟩ is used in English for /ʒ/ in words such as fusion.
⟨sk⟩ is used in Swedish to write the sje sound /ɧ/. It takes by rule this sound value before the front vowels (e, i, y, ä and ö) word or root initially (as in sked (spoon)), while normally representing /sk/ in other positions. In Norwegian and Faroese, it is used to write voiceless postalveolar fricative /ʃ/ (only in front of i, y, ei and øy/oy).
⟨sp⟩ is used in German for /ʃp/ as in Spaß /ʃpaːs/ instead of using schp (or chp).
⟨sr⟩ is used in Kosraean for /ʂ/.
⟨ss⟩ is used in Pinyin for /z/ in languages such as Yi. For its use in the Wade–Giles system of Romanization of Chinese, see Wade–Giles → Empty rime. In other languages, such as Catalan, Cornish, French, Italian, Occitan, Portuguese and Central Alaskan Yup'ik, where ⟨s⟩ transcribes /z/ between vowels (and elsewhere in the case of Yup'ik), ⟨ss⟩ is used for /s/ in that position (/sː/ in Italian and also in some cases in Cornish). In romanized Korean, it represents the fortis sound /s͈/.
⟨st⟩ is used in German for /ʃt/ as in Stadt /ʃtat/ instead of using scht (or cht). In some parts of northern Germany, the pronunciation /st/ (as in English) is still quite common in the local dialect.
⟨sz⟩ is used in several languages. See article.
⟨s-c⟩ and ⟨s-cc⟩ are used in Piedmontese for the sequence /stʃ/.
⟨s-g⟩ and ⟨s-gg⟩ are used in Piedmontese for the sequence /zdʒ/.
⟨ti⟩, before a vowel, is usually pronounced /sj/ in French.
⟨tj⟩ is used in Norwegian and Faroese words like tjære/tjøra ('tar') for /ç/ (Norwegian) and /tʃ/ (Faroese). In the closely related Swedish alphabet, it represents /ɕ/, as in tjära /ˈɕæːɾa/. It is, or was, also used for /tʃ/ in many Dutch-based orthographies in Indonesia and Surinam. In the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara, it represents a postalveolar stop, written /ṯ/ or /ḏ/. This sound is also written ⟨dj⟩, ⟨ty⟩, ⟨dy⟩, ⟨c⟩, or ⟨j⟩. In Catalan spelling it represents /d͡ʒ/
⟨tl⟩ is used in various orthographies for the affricate /tɬ/.
⟨tł⟩ is used in the transcription of Athabascan languages for a lateral affricate /tɬ/ or /tɬʰ/.
⟨tr⟩ generally represents a sound like a retroflex version of English "ch" in areas of German influence, such as Truk lagoon, now spelled ⟨chuuk⟩. For instance, in the orthography of Malagasy it represents /tʂ/. In southern dialects of Vietnamese, ⟨tr⟩ represents a voiceless retroflex affricate /tʂ/. In the northern dialects, this sound is pronounced /tɕ/, just like what ⟨ch⟩ represents. ⟨Tr⟩ was formerly considered a distinct letter of the Vietnamese alphabet, but today is not.
⟨ts⟩ is used in the orthography of Basque, where it represents an apical voiceless alveolar affricate /t̺s̺/. It contrasts with ⟨tz⟩, which is laminal /t̻s̻/. In the orthography of Hausa, ⟨ts⟩ represents an alveolar ejective fricative /sʼ/ or affricate /tsʼ/), depending on dialect. It is considered a distinct letter, and placed between ⟨t⟩ and ⟨u⟩ in alphabetical order. It is also used in the Catalan spelling for /t͡s/
The Wade-Giles and Yale romanizations of Chinese use ⟨ts⟩ for an unaspirated voiceless alveolar affricate /ts/. Wade-Giles also uses ⟨ts'⟩ for the aspirated equivalent /tsʰ/. These are equivalent to Pinyin ⟨z⟩ and ⟨c⟩, respectively. The Hepburn romanization of Japanese uses ⟨ts⟩ for a voiceless alveolar affricate /ts/). In native Japanese words, this sound only occurs before ⟨u⟩, but it may occur before other vowels in loanwords. Other romanization systems write /tsu/ as ⟨tu⟩. ⟨Ts⟩ in the orthography of Tagalog is used for /tʃ/. The sequence ⟨ts⟩ occurs in English, but it has no special function and simply represents a sequence of ⟨t⟩ and ⟨s⟩. It occurs word-initially only in some loanwords, such as tsunami and tsar. Most English-speakers do not pronounce a /t/ in such words and pronounce them as if they were spelled ⟨sunami⟩ and ⟨sar⟩, respectively.
⟨tt⟩ is used in the orthography of Basque for /c/, and in romanized Kabyle for /ts/. In romanized Korean, it represents the fortis sound /t͈/, and in Haida (Bringhurst orthography) it is ejective /tʼ/.
⟨tw⟩ is used for /tʷ/ in the orthography of Arrernte.
⟨tx⟩ is used in the orthographies of Basque, Catalan, as well as some indigenous languages of South America, for a voiceless postalveolar affricate /t͡ʃ/. In the orthography of Nambikwara it represents a glottalized /tʔ/.
⟨ty⟩ is used in the Hungarian alphabet for /cç/, a voiceless palatal affricate; in Hungarian, digraphs are considered single letters, and acronyms keep them intact. In the orthography of Xhosa, ⟨ty⟩ represents /tʲʼ/. In that of Shona, it represents /tʃk/. In the orthography of Tagalog it uses /tʃ/. In the transcription of Australian Aboriginal languages such as Warlpiri, Arrernte, and Pitjantjatjara, it represents a postalveolar stop, either voiceless /ṯ/ or voiced /ḏ/. (This sound is also written ⟨tj⟩, ⟨dj⟩, ⟨dy⟩, ⟨c⟩, and ⟨j⟩).
⟨tz⟩ is used in the orthographies of Basque and German for the voiceless alveolar affricate /t͡s/). In Basque, this sound is laminal and contrasts with the apical affricate represented by ⟨ts⟩. It is also used in Catalan to represent the voiced alveolar affricate /d͡z/. For its use in the Wade–Giles system of Romanization of Chinese, see Wade–Giles → Empty rime.
⟨uc⟩ is used in Nahuatl for /kʷ/ before a consonant. Before a vowel, ⟨cu⟩ is used.
⟨ue⟩ is found in many languages. In English, ⟨ue⟩ represents /ju/ or /u/ as in cue or true, respectively. In German, it is equivalent to Ü, and as such may appear in proper names of people, representing /ʏ/ or /yː/.
⟨ug⟩ is used in Central Alaskan Yup'ik for /ɣʷ/.
⟨ui⟩ in Dutch stands for the diphthong /œy/. In Irish and Scottish Gaelic, it is /ɪ/ after a velarized (broad) consonant, and in Irish, it is used for /ɪ/ /ʊ/ /iː/ /uː/ between a broad and a slender consonant. In German, it represents the diphthong /ʊɪ̯/, which appears only in interjections such as "pfui!". In English, it represents the sound /uː/ in fruit, juice, suit and pursuit. However, in many English words, this does not hold. For example, it fails in words where the u in ui functions as a modifier of a preceding g (forcing g to remain /ɡ/ rather than shifting to /dʒ/ in guild, guilt, guilty, sanguine, Guinea, etc.), doing the same with c (in words like circuit and biscuit), or in cases of unusual etymological spelling or syllable separation (e.g. build, suite, and intuition). In Mandarin pinyin, it is /wei̯/ after a consonant. (In initial position, this is spelled wei.) In French, it is not a digraph, but a predictable sequence /ɥi/, as in huit "eight".
⟨uí⟩ is used in Irish for /iː/ between a broad and a slender consonant.
⟨úi⟩ is used in Irish for /uː/ between a broad and a slender consonant.
⟨úm⟩ is used in Portuguese orthography for /ũ/ before a consonant.
⟨un⟩ is used in many languages to write a nasal vowel. In Portuguese orthography before a consonant, and in many West African languages, it is /ũ/, while in French it is /œ̃/, or among the younger generation /ɛ̃/. In pinyin, /u̯ən/ is spelled un after a consonant, wen initially.
⟨ún⟩ is used in Portuguese orthography for /ũ/ before a consonant.
⟨ün⟩ is used in Tibetan Pinyin for /ỹ/.
⟨uŋ⟩ is used in Lakhota for the nasal vowel /ũ/.
⟨uu⟩ is used in Dutch for /y/. In languages with phonemic long vowels, it may be used to write /uː/.
⟨vv⟩ is used in Central Alaskan Yup'ik for /f/.
⟨wh⟩ is used in English language to represent Proto-Germanic /hw/, the continuation of the PIE labiovelar */kʷ/; in Latin and Romance languages this instead became qu, as reflected in the English phonemic spelling kw (as in kwik). Most English interrogative words begin with this digraph, hence their name "wh-words", and indeed how began with the same sound in Proto-Germanic, but underwent a sound change (wh-cluster reductions) in Old English, hence the different spelling. This was formerly spelt ⟨hw⟩; spelling changed from hw to wh in Middle English, but pronunciation did not change. Pronunciation has changed over time; see phonological history of wh for details. In most dialects this is pronounced /w/ or /h/, following another wh-cluster reduction (/hw/ has merged with /w/ in most varieties of English in the wine–whine merger): /hw/ > /w/, or /hw/ > /h/ before /o/, as in who or whole (compare how, whose sound changed earlier and thus did not acquire the wh spelling). In the Māori language, ⟨wh⟩ represents /ɸ/ or more commonly /f/, with some regional variations approaching /h/ or /hw/. In the Taranaki region, for some speakers, this represents a glottalized /wʼ/. In Xhosa, it represents /w̤/, a murmured variant of /w/ found in loan words. In Cornish, it represents /ʍ/.
⟨wr⟩ is now used by most English dialects for /r/. It once was not a digraph but represented the predictable sequence /wr/, a value it retains in a few dialects documented in the twentieth century.
⟨ww⟩ is used in Haida (Bringhurst orthography) for glottalized /ˀw/.
⟨xh⟩, in Albanian, represents the sound of the voiced postalveolar affricate consonant /dʒ/, as in the surname Hoxha /ˈhɔdʒa/. In Pashto too it represents /dʒ/. In Zulu and Xhosa it represents the voiceless aspirated alveolar lateral click /kǁʰ/, for example in the name of the language Xhosa /ˈkǁʰoːsa/. In Walloon to write a sound that is variously /h/ or /ʃ/, depending on the dialect. In Canadian Tlingit it represents /χ/, which in Alaska is written x̱.
⟨xi⟩ is used in English for /kʃ/ in words such as flexion. (It is equivalent to ⟨c⟩ plus the digraph ⟨ti⟩, as in action.)
⟨xu⟩ was used in the Ossete Latin alphabet for /χʷ/.
⟨xw⟩ is used in the Tlingit language for /xʷ/.
⟨xx⟩ is used in Hadza for the glottalized click /ᵑǁˀ/.
⟨yh⟩ was used in the pre-1985 orthography of Guinea, for the "ejective y" or palatalized glottal stop (/ʔʲ/) in Pular (a Fula language). In the current orthography it is now written ƴ. In Xhosa it is used for the sound / j̈ /. In a handful of Australian languages, it represents a "dental semivowel".[clarification needed]
⟨yi⟩ is used in Mandarin pinyin to write the vowel /i/ when it forms an entire syllable.
⟨ym⟩ is used in French to write the vowel sound /ɛ̃/ (/im/ before another vowel), as in thym /tɛ̃/ "thyme".
⟨yn⟩ is used in French to write the vowel sound /ɛ̃/ in some words of Greek origin, such as syncope /sɛ̃kɔp/ "syncope".
⟨yu⟩ is used in romanized Chinese to write the vowel /y/. In Mandarin pinyin it is used for /y/ in initial position, whereas in Cantonese Jyutping it is used for /y/ in non-initial position. (See jyu.)
⟨yy⟩ is used in some languages such as Finnish to write the long vowel /yː/. In Haida (Bringhurst orthography) it is glottalized /ˀj/.
⟨zh⟩ represents the voiced postalveolar fricative (/ʒ/), like the ⟨s⟩ in pleasure, in Albanian and in Native American orthographies such as Navajo. It is used for the same sound in some English-language dictionaries, as well as to transliterate the sound when represented by Cyrillic ⟨ж⟩ and Persian ⟨ژ⟩ into English; though it is rarely used for this sound in native English words (perhaps the only one being zhoosh). ⟨Zh⟩ as a digraph is rare in European languages using the Latin alphabet; in addition to Albanian it is found in Breton in words that are pronounced with /z/ in some dialects and /h/ in others. In Hanyu Pinyin, ⟨zh⟩ represents the voiceless retroflex affricate /tʂ/. When the Tamil language is transliterated into the Latin script, ⟨zh⟩ represents a retroflex approximant (Tamil ழ U+0BB4, ḻ, [ɹ]).
⟨zs⟩ is the last (forty-fourth) letter of the Hungarian alphabet. Its name is "zsé" and represents /ʒ/, a voiced postalveolar fricative, similar to J in Jacques and s in vision. A few examples are rózsa "rose" and zsír "fat".
⟨ǀn⟩ ⟨ǁn⟩ ⟨ǃn⟩ ⟨ǂn⟩ are used in Nama for its four plain nasal clicks.
- Chubb, Ray (2013) [First published 2010]. "Leveryans – Pronunciation". Skeul an Tavas: A coursebook in Standard Cornish. Illustrations by Nigel Roberts (Second ed.). Cnoc Sceichín, Leac an Anfa, Cathair na Mart, Co. Mhaigh Eo: Evertype. pp. 84–94. ISBN 978-1-904808-93-0.
- Chubb, Ray (2011) [First published 2010]. "Leveryans – Pronunciation". Skeul an Tavas: A Cornish language coursebook for adults in the Standard Written Form with Traditional Graphs. Illustrations by Nigel Roberts (Second ed.). Redruth, Kernow / Cornwall, UK: Agan Tavas. pp. 84–92. ISBN 978-1-901409-12-3.
- Chubb, Ray (2013) [First published 2010]. "Leveryans – Pronunciation". Skeul an Tavas: A Cornish language coursebook for schools in the Standard Written Form. Illustrations by Nigel Roberts (Second ed.). Redruth, Kernow / Cornwall, UK: Agan Tavas. pp. 84–92. ISBN 978-1-901409-13-0.
- Bock, Albert; Bruch, Benjamin (3 July 2008). "An Outline of the Standard Written Form of Cornish" (First ed.). ISBN 978-1-903798-56-0. Retrieved 6 December 2014.
- Rickard, Peter (2000). A history of the French language (2. ed., reprinted. ed.). London: Routledge. p. 22. ISBN 0-415-10887-X.
- First Lt. William E. W. MacKinlay, 1905, A Handbook and Grammar of the Tagalog Language. Washington: Government Printing Office.
- Edward von Preissig, 1918, Dictionary and Grammar of the Chamorro Language of the Island of Guam. Washington: Government Printing Office.
- "L’orthographe des langues de la République démocratique du Congo: entre usages et norme". Les cahiers du Rifal 23.
- George, Ken, ed. (September 2009) [First edition published in 1993 under the title Gerlyver Kernewek Kemmyn – An Gerlyver Meur]. "6. Recommended pronunciation". Gerlyver Meur (Second ed.). Cornish Language Board. pp. 28–35. ISBN 978-1-902917-84-9.
- IPA: Vowels
- 董峰政, "Taiwanese Tong-iong Pingim Dictionary", 臺南市寧南語言文化協會, Tainan City,Jul 2006.
- Williams, Nicholas (2006). "Pronunciation and Spelling of Unified Cornish Revised". In Everson, Michael. English–Cornish Dictionary: Gerlyver Sawsnek–Kernowek (Second ed.). Redruth, Kernow, UK: Agan Tavas. pp. xxvii–xxx. ISBN 978-1-901409-09-3.