View source for Silent letter

{{short description|Letter that is not pronounced}}
{{About|a feature in alphabetic writing systems|the album by America|Silent Letter}}
{{more citations needed|date=September 2014}}
{{IPA notice}}
In an [[alphabet|alphabetic writing system]], a '''silent letter''' is a [[Letter (alphabet)|letter]] that, in a particular [[word]], does not correspond to any [[Phone (phonetics)|sound]] in the word's [[pronunciation]]. In linguistics, a silent letter is often symbolised with a [[null sign]] {{unichar|2205|Empty Set}}, which resembles the [[Ø|Scandinavian letter Ø]]. A [[Zero (linguistics)|null]] or zero is an unpronounced or unwritten segment.

==English==
{{unreferenced section|date=December 2018}}
{{See also|English orthography}}
{{for|historical explanations for some silent letters in English|Phonological history of English consonant clusters}}
One of the noted difficulties of [[English spelling]] is a high number of silent letters. [[Faculty of Humanities (University of Manchester)|Edward Carney]] distinguishes different kinds of "silent" letters, which present differing degrees of difficulty to readers.
* '''Auxiliary letters''' which, with another letter, constitute [[digraph (orthography)|digraph]]s; i.e., two letters combined which represent a single [[phoneme]]. These may further be categorized as:
** "Exocentric" digraphs, where the sound of the digraph is different from that of either of its constituent letters. These are rarely considered "silent". Examples:
*** Where the phoneme has no standard single-letter representation, as with [[consonant]]s {{angbr|ng}} for {{IPA|/ŋ/}} as in ''sing'', {{angbr|th}} for {{IPA|/θ/}} as in ''thin'' or {{IPA|/ð/}} as in ''then'', or [[diphthong]]s {{angbr|ou}} in ''out'' or {{angbr|oi}} in ''point''. These are the default spellings for the relevant sounds and present no special difficulty for readers or writers.
*** Where standard single-letter representation uses another letter, as with {{angbr|gh}} in ''enough'' or {{angbr|ph}} in ''physical'' instead of {{angbr|f}}. These may be considered irregular for writers, but less difficult for readers.
** "Endocentric" digraphs, where the sound of the digraph is the same as that of one of its constituent letters. These include:
*** Most [[Digraph (orthography)#Double letters|double consonants]], as {{angbr|bb}} in ''clubbed''; though not [[geminate consonant]]s, as {{angbr|ss}} in ''misspell''. Doubling due to [[suffix]]ation or [[inflection]] is regular; otherwise,{{clarify|date=October 2022}} it may present difficulty to writers (e.g., ''accommodate'' is often misspelled), but not to readers.
*** Many vowel digraphs, as {{angbr|ea}}, {{angbr|ie}}, {{angbr|eu}} in ''leave'' (cf. ''accede''), ''achieve'', ''eulogy'' (cf. ''utopia'').
*** The discontiguous digraphs, whose second element is "[[Silent e|magic e]]"; e.g., {{angbr|a_e}} in ''rate'' (cf. ''rat''), {{angbr|i_e}} in ''fine'' (cf. ''fin''). This is the regular way to represent "long" [[vowel]]s in the last [[syllable]] of a [[morpheme]].
*** Others, such as {{angbr|ck}} (which is in effect the "doubled" form of {{angbr|k}}), {{angbr|gu}} as in ''guard'', ''vogue''; {{angbr|ea}} as in ''bread'', ''heavy'', etc.; {{angbr|ae}}, {{angbr|oe}} as in ''aerial'', ''oedipal''. These may be difficult for writers and sometimes also for readers.
* '''Dummy letters''' with no relation to neighboring letters and no correspondence in pronunciation:
** Some are '''inert letters''', which are sounded in a [[cognate]] word; e.g., {{angbr|n}} in ''damn'' (cf. ''damnation''); {{angbr|g}} in ''phlegm'' (cf. ''phlegmatic''); {{angbr|a}} in ''practically'' (cf. ''practical''); {{angbr|t}} in ''ballet'' (cf. ''balletic''); {{angbr|b}} in su''b''tle (cf. ''subtility''). If the cognate is obvious, it may aid writers in spelling, but mislead readers in pronunciation.
** The rest are '''empty letters''', which never have a sound; e.g., {{angbr|b}} in dou''b''t, {{angbr|h}} in ''h''onor, {{angbr|w}} in ans''w''er, {{angbr|h}} in Sara''h'', {{angbr|s}} in i''s''land. These may present the greatest difficulty to writers and often to readers, as well.

The distinction between "endocentric" digraphs and empty letters is somewhat [[arbitrary]]. For example, in such words as ''little'' and ''bottle'', one might view {{angbr|le}} as an "endocentric" digraph for {{IPA|/əl/}}, or view {{angbr|e}} as an empty letter; similarly, with {{angbr|bu}} or {{angbr|u}} in ''buy'' and ''build''.

Not all silent letters are completely redundant:
* Silent letters can distinguish between [[homophone]]s; e.g., ''in''/''inn''; ''be''/''bee''; ''lent''/''leant''. This is an aid to readers already familiar with both words.
* Silent letters may give an insight into the [[Meaning (linguistics)|meaning]] or [[Etymology|origin of a word]]; e.g., ''vineyard'' suggests [[vine]]s more than the phonetic ''*vinyard'' would.
* Silent letters may help the reader to [[Stress (linguistics)|stress]] the correct [[syllable]] (compare ''physics'' to ''physiques''). The final {{angbr|fe}} in ''giraffe'' gives a clue to the second-syllable stress, where ''*giraf'' might suggest initial-stress.

Silent letters arise in several ways:
* [[Sound change]]s occurring without a spelling change. The [[digraph (orthography)|digraph]] {{angbr|gh}} [[Gh (digraph)#English|was pronounced]] {{IPAblink|x}} in [[Middle English language|Middle English]] in such words as ''light''.
* Sound distinctions from [[foreign language]]s may be lost, as with the distinction between smooth [[Rho (letter)|rho]] (ρ) and roughly [[aspirated consonant|aspirated]] rho (ῥ) in [[Greek language|Ancient Greek]], represented by {{angbr|r}} and {{angbr|rh}} in [[Latin]], but merged to the same {{IPA|[r]}} in [[English language|English]].
* [[English consonant-cluster reductions|Clusters of consonants may be simplified]], producing silent letters; e.g., silent {{angbr|th}} in ''asthma'', silent {{angbr|t}} in ''Christmas'' (in conservative [[Received Pronunciation|RP]], it is pronounced /krɪstməs/, as opposed to /krɪsməs/ in all other dialects). Similarly, with alien clusters, such as [[Greek language|Greek]] initial {{angbr|ps}} in ''psychology'' and {{angbr|mn}} in ''mnemonic'', and the much rarer clusters in ''[[chthonic]]'' and ''[[phthalate]]''.
* [[Compound word]]s are often simplified in pronunciation, while their spelling remains the same. For example, ''cupboard'' and ''breakfast'' were once pronounced as written, but were then simplified over time. The words ''forehead'' and ''waistcoat'' have largely reverted to their [[spelling pronunciation]]s, but were once pronounced *''forrid'' and *''weskit'', respectively.
* Occasionally, spurious letters are consciously inserted in spelling to reflect [[etymology]] (real or imagined). The {{angbr|b}} in ''debt'' and ''doubt'' (from [[French language|French]] ''dette, doute'') was inserted to match [[Latin]] cognates like ''debit'' and ''dubitable''. A silent {{angbr|s}} was inserted in ''isle'' ([[Norman French]] ''ile'', [[Old French]] ''isle'', from Latin ''insula''; cognate to ''isolate'') and then extended to the unrelated word ''island''. The {{angbr|p}} in ''ptarmigan'' was apparently suggested by Greek words such as ''pteron'' ('wing').

Since [[Accent (sociolinguistics)|accent]] and pronunciation differ, letters may be silent for some speakers, but not others. In [[non-rhotic]] accents, {{angbr|r}} is silent in such words as ''hard'', ''feathered''; in [[h-dropping|''h''-dropping]] accents, {{angbr|h}} is silent. A speaker may or may not pronounce {{angbr|t}} in ''often'', the first {{angbr|c}} in ''Antarctic'', {{angbr|d}} in ''sandwich'', etc.

===Differences between British English and American English===

====Pronunciation====

In the US, the ''h'' in ''herb'' is silent (''a'''n''' herb''), but in the UK, it is pronounced (''a herb''). The same is true for the ''l'' in ''solder.''

In parts of the UK, the ''a'' in ''dictionary'' and ''secretary'' is silent, but in the US, it is pronounced.

====Spelling====

In US spellings, silent letters are sometimes omitted (e.g., ''acknowledgment'' / UK ''acknowledg'''e'''ment'', ''ax'' / UK ''ax'''e''''', ''catalog'' / UK ''catalog'''ue''''', ''program'' / UK ''program'''me''''' outside computer contexts), but not always (e.g., ''dialog'''ue''''' is the standard spelling in the US and the UK; ''dialog'' is regarded as a US variant; the spelling ''axe'' is also often used in the US). In most words, silent letters are written in both styles (e.g., ''de'''b'''t'', ''g'''u'''ard'', ''hous'''e''''').

==Other Germanic languages==

===Danish===
The [[Danish language]] has different letters that can be silent:

* The letter {{angbr|f}} is silent in the preposition ''af'' ('by, of, from, off, with, out of').
* The letter {{angbr|g}} is silent in the conjunction ''og'' ('and') and adverb ''også'' ('also').
* The letter {{angbr|h}} is silent in most dialects if followed by {{angbr|v}}, as in the pronouns ''hvad'' ('what'), ''hvem'' ('who'), ''hvor'' ('where').<ref>[[:da:Stumt bogstav]]</ref>
* The letter {{angbr|v}} is silent at the end of words if preceded by {{angbr|l}}, as in the pronoun ''selv'' ('self') and adjective ''halv'' ('half').
* The letter {{angbr|d}} is usually (but not necessarily) silent if preceded by a consonant, as in ''en mand'' ('a man') and ''blind'' ('blind'). Many words ending in {{angbr|d}} are pronounced with a [[stød]], but it is still considered a silent letter.<ref>{{Cite web|url=http://www.denstoredanske.dk/Sprog,_religion_og_filosofi/Sprog/Ortografi/d_D|title = D, d – bogstav &#124; lex.dk| date=20 April 2023}}</ref>

===Faroese===
{{unreferenced section|date=December 2018}}
The [[Faroese language]] has two silent letters.

The letter {{lang|fo|edd}} {{angbr|ð}} is almost always silent. It is rendered in orthography for historical reasons (e.g., {{lang|fo|faðir}} 'father' {{IPA|fo|ˈfɛajɪɹ|}}, cf. [[Old Norse]] {{lang|non|faðir}}). In some cases, however, the letter {{lang|fo|edd}} is pronounced {{IPA|fo|ɡ̊|}}, as in {{lang|fo|veðrið}} 'the weather' {{IPA|fo|ˈvɛɡ̊ʐɪ|}}.

The letter {{lang|fo|ge}} {{angbr|g}} (i.e. continuant of [[Old Norse]] {{IPA|non|ɣ|}}) is usually silent between vowels or when following a vowel before a pause (e.g., {{lang|fo|dagur}} 'day' {{IPA|fo|ˈd̥ɛavʊɹ|}}, cf. [[Old Norse]] {{lang|non|dagr}} {{IPA|non|ˈdaɣʐ|}}; {{lang|fo|eg}} 'I' {{IPA|fo|ˈeː|}}, cf. [[Old Norse]] {{lang|non|ek}}). Use of the silent letter {{lang|fo|ge}} in Faroese is the same as for the letter ''edd''; it is written for historical reasons as [[Faroese orthography]] was based on normalised spelling of [[Old Norse]] and [[Icelandic language]].

Both Faroese silent letters {{lang|fo|edd}} and {{lang|fo|ge}} are replaced by a [[Hiatus (linguistics)|hiatus]] [[Faroese orthography#Glide insertion|glide consonant]] ({{IPA|fo|j|}}, {{IPA|fo|v|}} or {{IPA|fo|w|}}) when followed by another (unstressed) vowel.

===German===
In German, silent letters are rare apart from word-internal {{angle bracket|h}} (following a vowel) and the {{angle bracket|e}} in the [[digraph (orthography)|digraph]] {{angle bracket|ie}}.

===={{angle bracket|h}}====
Silent ''h'' is used in German to indicate vowel length or [[hiatus (linguistics)|hiatus]]. This ''h'' is almost regularly added at the end of [[inflection|inflectable]] word stems, e.g. ''Ku'''h''''' (cow), ''Stro'''h''''' (straw), ''dre'''h'''en'' (to turn, stem ''dreh''-). There is only a fairly small number of exceptions to this, mostly nouns in ''-ee'' or ''-ie'' (see below), apart from isolated cases such as ''säen'' (to sow).

Otherwise silent ''h'' may be written before the letters ''l, m, n, r'' as in ''ne'''h'''men'' (to take), ''Stu'''h'''l'' (chair), ''Za'''h'''n'' (tooth). This latter use is highly irregular, however, and there are just as many words where the ''h'' is missing.

Historically, this use of silent ''h'' goes back to the [[Middle High German]] consonant {{IPA|/h/}}, which became silent in words like ''se'''h'''en'' (to see), ''ze'''h'''n'' (ten). By analogy it was then also used in words that had no such ''h'' in Middle High German. The majority of silent ''h''{{'s}} in modern German are analogical rather than etymological.

===={{angle bracket|e}}====
The long ''i''-sound {{IPAslink|iː}} is usually written {{angle bracket|ie}}, with a silent {{angle bracket|e}}, as in ''v'''ie'''l'' (much), ''sp'''ie'''len'' (to play), ''W'''ie'''n'' ('[[Vienna]]'), and hundreds of other words.

In native German words this spelling is fairly unambiguous. Some words of foreign origin also behave like native words; e.g., ''Kur'''ie'''r, Pap'''ie'''r, Turn'''ie'''r'' and all verbs ending in {{angle bracket|-ieren}} (e.g. ''appell'''ie'''ren, organis'''ie'''ren''). In other foreign words, however, the {{angle bracket|e}} after {{angle bracket|i}} may be pronounced (e.g., ''Ambi'''e'''nte'', ''Hygi'''e'''ne'', ''Kli'''e'''nt''), or names like ''Dani'''e'''la'', ''Gabri'''e'''l'', and ''Tri'''e'''st''.

Words ending in {{angle bracket|-ie}} can be particularly tricky to learners: There are generally two possibilities:

# When the final {{angle bracket|ie}} is stressed, it represents long {{IPA|/iː/}} as in ''Zeremon'''ie''''' {{IPA|/tseʁemoˈniː/}}. Some other words with {{angle bracket|ie}} pronounced this way include ''Akadem'''ie''''', ''Allerg'''ie''''', ''Amnes'''ie''''', ''Amnest'''ie''''', ''Apath'''ie''''', ''Artiller'''ie''''', ''Batter'''ie''''', ''Blasphem'''ie''''', ''Chem'''ie''''', ''Chirurg'''ie''''', ''Demokrat'''ie''''', ''Energ'''ie''''', ''Epidem'''ie''''', ''-gam'''ie''''', ''Garant'''ie''''', ''Gen'''ie''''', ''Geometr'''ie''''', ''-graf'''ie'''/-graph'''ie''''', ''Harmon'''ie''''', ''Hyster'''ie''''', ''Infanter'''ie''''', ''Iron'''ie''''', ''Kavaller'''ie''''', ''Kompan'''ie''''', ''Kop'''ie''''', ''[[wikt:-logie#German|-log'''ie''']]'', ''Liturg'''ie''''',  ''Mag'''ie''''', ''Man'''ie''''', ''Mar'''ie''''', ''Melod'''ie''''', ''Monoton'''ie''''', ''Nostalg'''ie''''', ''Orthopäd'''ie''''', ''Part'''ie''''', ''Phantas'''ie''''', ''Philosoph'''ie''''', ''Poes'''ie''''', ''Psychiatr'''ie''''', ''Rhapsod'''ie''''', ''Sinfon'''ie''''', ''-skop'''ie''''', ''Theor'''ie''''', ''Therap'''ie''''', and ''Utop'''ie'''''. 
# When the preceding vowel is stressed, {{angle bracket|ie}} represents the separate vowels {{IPA|/i.ə/}} as in ''Foli'''e''''' {{IPA|/ˈfoːliə/}}. Some other words with {{angle bracket|ie}} pronounced this way include ''Akazi'''e''''', ''Akti'''e''''', ''Amali'''e''''', ''Begoni'''e''''', ''Emili'''e''''', ''Famili'''e''''', ''Foli'''e''''', ''Gerani'''e''''', ''Grazi'''e''''', ''Hortensi'''e''''', ''Hosti'''e''''', ''Immobili'''e''''', ''Kastani'''e''''', ''Komödi'''e''''', ''Kuri'''e''''', ''Lili'''e''''', ''Lini'''e''''', ''Orgi'''e''''', ''Otili'''e''''', ''Pini'''e''''',  ''Seri'''e''''', ''Studi'''e''''', ''Tragödi'''e''''', and ''Zäzili'''e'''''. 
# In female names, there is a third category of words stressed on the [[antepenult]], where {{angle bracket|ie}} also represents {{IPA|/iː/}}; e.g., ''Amel'''ie''''', ''Leon'''ie''''', ''Nathali'''e''''', ''Rosal'''ie''''', ''Stefan'''ie''''', ''Valer'''ie''''' (all stressed on the first syllable).

A special case arises when the {{angle bracket|e}} after {{angle bracket|i}} is a grammatical ending; in this case, it is always pronounced. Therefore, ''Zeremon'''ie''''' becomes ''Zeremoni'''e'''n'' {{IPA|/tseʁemoˈniːən/}} in the plural,<ref>{{cite web |url=http://de.pons.com/übersetzung?q=Zeremonie&l=deen&in=de&lf=de |title=Zeremonie |author=<!--Staff writer(s); no by-line.--> |publisher=PONS |access-date=12 October 2015}}</ref> and the same is true of all other nouns in group 1 above. The noun ''Knie'' is pronounced {{IPA|/kniː/}} when it is singular, but usually {{IPA|/ˈkniːə/}} when it is plural. ''Spermi'''e'''n'' is plural of ''Spermium'', hence also with a pronounced {{angle bracket|e}}. Country names in ''-ien'' can also be joined to this group: ''Australi'''e'''n'', ''Brasili'''e'''n'', ''Indi'''e'''n'', ''Kroati'''e'''n'', ''Serbi'''e'''n'', ''Sloweni'''e'''n''.

====Other letters====
Other silent letters occur mainly in borrowings from [[French language|French]] and other modern languages; e.g., ''Porträ'''t''''' (portrait), ''Kor'''ps''''' (corps).

Informally, the letter {{angle bracket|t}} may be silent in [[function word]]s like ''is'''t''''' (is), ''jetz'''t''''' (now), ''nich'''t''''' (not), and otherwise in clusters like ''Gedäch'''t'''nis'' (memory), ''Kuns'''t'''stück'' (piece of art). These ''t''{{'s}} are commonly silent in everyday speech, but will be retained in careful, formal parlance.

==Romance languages==

===French===

Silent letters are common in [[French language|French]], including the last letter of most words. Ignoring auxiliary letters that create digraphs (such as {{angbr|ch}}, {{angbr|gn}}, {{angbr|ph}}, {{angbr|au}}, {{angbr|eu}}, {{angbr|ei}}, and {{angbr|ou}}, as well as {{angbr|m}} and {{angbr|n}} as signals for [[nasalized vowel]]s), they include almost every possible letter except {{angbr|j}} and {{angbr|v}}.{{Citation needed|date=March 2024}}

====Vowels====
Final {{angbr|e}} is silent or at least (in poetry and song) a nearly-silent [[schwa]] {{IPA|/ə/}}; it allows the preservation of a preceding consonant, often allowing the preservation of a grammatical distinction between [[Grammatical gender|masculine and feminine]] forms in writing; e.g., in ''vert'' and ''verte'' (both 'green'); the {{angbr|t}} is pronounced in the latter (feminine) but not the former. Furthermore, the schwa can prevent an awkward ending of a word ending in a consonant and a liquid (''peuple, sucre'').

After {{angbr|é}}, {{angbr|i}}, or {{angbr|u}}, a final {{angbr|e}} is silent. The spelling {{angbr|eau}} is pronounced just the same as that for {{angbr|au}} and is entirely an etymological distinction, so in that context, the {{angbr|e}} is silent.

The digraph {{angbr|qu}} for {{IPA|/k/}} usually has a silent {{angbr|u}}, as in ''quand'' ('when'), ''quel'' ('which'), ''acquérir'' ('to acquire'), and ''quotidien'' ('daily'). {{angbr|gu}} for {{IPA|/ɡ/}} has the same silent {{angbr|u}}; when the {{angbr|u}} is not silent, it is usually marked with a [[trema (diacritic)|trema]]: {{angbr|ü}}.

====Consonants====
{{angbr|h}} is silent outside of the digraph {{angbr|ch}}. Numerous doubled consonants exist; French does not distinguish doubled consonants from single consonants in pronunciation as [[Italian language|Italian]] does. A marked distinction exists between a single and doubled {{angbr|s}}: doubled {{angbr|ss}} is always [[voiceless]] {{IPAblink|s}}, while an intervocalic single {{angbr|s}} is usually [[voiced]] {{IPAblink|z}}.

The nasal consonants {{angbr|[[m]]}} and {{angbr|[[n]]}} when final or preceding a consonant ordinarily nasalize a preceding vowel but are not themselves pronounced (''faim'', ''tomber'', ''vin'', ''vendre''). Initial and intervocalic {{angbr|m}} and {{angbr|n}}, even before a final silent {{angbr|e}}, are pronounced: ''aimer'', ''jaune''.

Most final consonants are silent, except in most cases with the letters {{angbr|c}}, {{angbr|f}}, {{angbr|l}}, and {{angbr|r}} (the English word '''''c'''a'''r'''e'''f'''u'''l''''' is a [[mnemonic]] for this set). But even this rule has its exceptions: final morphemic {{angbr|er}} is usually pronounced /e/ (={{angbr|é}}) rather than the expected /ɛʁ/ (as in ''hiver'', which does not have that morpheme). Final {{angbr|l}} and {{angbr|ll}} is silent after {{angbr|i}} even in a diphthong (''œil'', ''appareil'', ''travail'', ''bouillir''). Final -''ent'' is silent as a third-person plural verb ending, though it is pronounced in other cases.

Final consonants that might be silent in other contexts (finally or before another consonant) may seem to reappear in pronunciation in [[liaison (French)|liaison]]: ''ils ont'' {{IPA|fr|ilz‿ɔ̃|}} "they have", as opposed to ''ils sont'' {{IPA|fr|il sɔ̃|}} "they are"; liaison is the retention (between words in certain syntactic relationships) of a [[historical linguistics|historical sound]] otherwise lost, and often has grammatical or lexical significance.

===Italian===
The letter {{angbr|h}} most often marks a {{angbr|c}}/{{angbr|g}} as hard ([[velar consonant|velar]]), as in ''spaghetti'' and ''scherzo'', where it would otherwise be soft ([[palatal consonant|palatal]]), as in ''gelato'' and ''violoncello'', because of a following [[front vowel]] ({{angbr|e}} or {{angbr|i}}).

Conversely, to soften {{angbr|c}} or {{angbr|g}} (to {{IPA|/tʃ/}} or {{IPA|/dʒ/}} respectively) before a [[back vowel]] ({{angbr|a}}, {{angbr|o}}, {{angbr|u}}), a silent {{angbr|i}} is inserted: {{angbr|–cio–}}, {{angbr|–giu–}}, etc. When {{angbr|i}} in that position is not silent, it can be marked with a grave accent: {{angbr|ì}}. Before any other letter, or at the end of a word, the {{angbr|i}} is ''not'' silent. {{see also|Hard and soft C|Hard and soft G}}

Silent {{angbr|h}} is also used in forms of the verb ''avere'' ('have') – ''ho'', ''hai'' and ''hanno'' – to distinguish these from their homophones ''o'' ('or'), ''ai'' ('to the') and ''anno'' ('year'). The letter {{angbr|h}} is also silent at the beginning of words borrowed from other languages, such as ''hotel''.

===Portuguese===

[[European Portuguese]]'s orthography used to conserve the etymological silent letters {{angbr|c}} and {{angbr|p}} when they appeared after a vowel and before the consonants {{angbr|t}}, {{angbr|ç}} or soft {{angbr|c}}. Their purpose was to prevent the preceding vowel's deafening, compare ''ad'''op'''tar'' {{IPA|fo|ɐdoˈtaɾ|}} (adopt), ''contr'''ac'''ção'' {{IPA|fo|kõtɾaˈsɐ̃w|}} (contraction) and ''insp'''ec'''cionar'' {{IPA|fo|ĩʃpesiuˈnaɾ|}} (inspect) to ''adj'''a'''cente'' {{IPA|fo|ɐdʒɐˈsẽtɨ|}} (adjacent), ''compl'''e'''tar'' {{IPA|fo|kõpɫɨˈtaɾ|}} (complete) and ''n'''o'''ção'' {{IPA|fo|nuˈsɐ̃w|}} (noction). However, the latest [[Portuguese Language Orthographic Agreement of 1990|Portuguese orthographic reform]]  removed them from the language entirely, meaning that enlarged vowels and deafened vowels cannot be distinguished solely by spelling anymore, the main reason for this change was that [[Brazilian Portuguese]]'s spelling had already stopped writing these mute consonants, and thus they were a figure of divergence between the two countries' way of writing.<ref name=":1">{{Cite web |title=Decreto nº 6.583, de 29 de setembro de 2008 |url=https://download.uol.com.br/educacao/UOL_Educacao_Integra_do_Acordo_Ortografico.pdf}}</ref>

Current Portuguese still has some mute letters:

====Vowels====
The {{angbr|u}} in the digraph {{angbr|gu}} is not pronounced before {{angbr|e}} and {{angbr|i}}, it is there because a single {{angbr|g}} before these vowels becomes a {{IPA|/ʒ/}}. There are exceptions to this rule, though, like the words ''aguentar'' and ''arguição''.

The {{angbr|u}} in the digraph {{angbr|qu}}, the latter is pronounced as {{IPA|/k/}} or as {{IPA|/kʷ/}}, is not pronounced before {{angbr|e}} and {{angbr|i}}, it is the only traditional way of putting a {{IPA|/k/}} sound before these letters, as the letter {{angbr|c}} softens to {{IPA|/s/}} and {{angbr|k}} is reserved to loanwords. There are exceptions to this rule, though, like the words ''eloquente'' and ''tranquilo''.

The {{angbr|u}} in the diphthong {{angbr|ou}} is not generally pronounced in [[European Portuguese|Standard Lisbon Portuguese]]<ref>{{Cite book |last=Cruz-Ferreira |first=Madalena |url=https://archive.org/details/rosettaproject_por_phon-1 |title=Handbook of the International Phonetic Association: A Guide to the Use of the International Phonetic Alphabet |publisher=Cambridge University Press |year=1999 |isbn=0-521-63751-1 |editor-link=International Phonetic Association |pages=126–130 |chapter=Portuguese (European)}}</ref> and some Brazilian dialects, which results in words like ''Sousa'' and ''Gouveia'' being said as ''"Sosa"'' and ''"Goveia"'' respectively.

====Consonants====
{{angbr|h}} is always silent at the beginning and at the ending of words (''hoje'', ''hora'', ''ah!'', etc.), except in loanwords such as ''hobby''. However, {{angbr|h}} is present in three digraphs and one trigraph (''ch'', ''lh'', ''nh'' and ''tch''), which is where this letter is mainly found affecting the word's pronunciation.

{{angbr|m}} and {{angbr|n}} are nasal consonants, when at the end of a word or behind another consonant, they nasalize the preceding vowel, with this being the most common way of encountering nasal vowels in Portuguese, the other being with the [[tilde]]. They are not interchangeable, though, {{angbr|m}} is only used before {{angbr|b}} and {{angbr|p}} (''embora'', ''império'', etc.) while {{angbr|n}} is used everywhere else (''canto'', ''circunstância'', ''convencer'', ''ênfase'', ''fundo'', ''lançar'', etc.). Other difference between these two letters is that paroxytones ending with an {{angbr|em}} are not accentuated while oxytones have to be accented, (''coragem'', ''também'', etc.), but the opposite happens with {{angbr|en}} (''pólen'', ''nenen'', etc.), all proparoxytones are accentuated in Portuguese.

{{angbr|x}} is normally deemed as silent when behind a soft {{angbr|c}} or the letter {{angbr|s}}, like in the words ''excelência'', ''excisar'' or ''exsudar'', but some people consider {{angbr|xc}} and {{angbr|xs}} to just be digraphs.

There is an extremely specific and limited group of words whose mute consonants can currently be preserved, but only when they occur at the end of them, these are onomastic forms in which the usage has consecrated them, namely anthroponyms and toponyms of biblical tradition like ''[[Jacob]]'', ''[[Job (biblical figure)|Job]]'', ''[[Isaac]]'', ''[[David]]'', ''[[Gad (son of Jacob)|Gad]]'', ''[[Gog and Magog|Gog]]'', ''[[Magog (Bible)|Magog]]'' and ''[[Jehoshaphat|Josafat]]''. Some others names, like ''[[Madrid]]'', ''[[Valladolid|Valhadolid]]'', ''[[Kozhikode|Calecut]]'' and ''[[Kozhikode|Calicut]]'' also apply to this rule. Although, all of the anthroponyms cited here can also be optionally written without them: ''[[Jacob|Jacó]]'', ''[[Job (biblical figure)|Jó]]'', ''[[David|Davi]]'', etc.<ref name=":1" />

===Spanish===

Despite being rather [[phonemic orthography|phonemic]], [[Spanish orthography]] retains some silent letters:
* {{angbr|h}} is silent outside of the digraph {{angbr|ch}} and loanwords such as ''hámster'' or ''[[wikt:hachís|hachís]]''.
* The digraph {{angbr|qu}}, used to represent {{IPAblink|k}} before the front vowels {{angbr|e}} and {{angbr|i}}, has a silent {{angbr|u}}. In contrast, the u is pronounced in {{angbr|cu}}.
* {{angbr|gu}} for {{IPA|/ɡ/}} has the same silent {{angbr|u}} before {{angbr|e}} and {{angbr|i}}. When the {{angbr|u}} is not silent, it must be marked with a [[trema (diacritic)|trema]]: {{angbr|ü}}. Before {{angbr|a}} and {{angbr|o}}, the {{angbr|u}} is ''not'' silent.
* Syllable onsets like {{angbr|pn}}, {{angbr|ps}}, {{angbr|pt}} are usually pronounced without ''p''. In some but not all cases, this is reflected in the orthography that allows double forms with and without {{angbr|p}}: {{lang|es|neumático}} ("tire"), {{lang|es|psicología}}/{{lang|es|sicología}} ("psychology"), {{lang|es|pterodáctilo}}/{{lang|es|terodáctilo}} ("[[pterodactyl]]"). Reducing an internal {{angbr|pt}} ({{lang|es|*conceto}} instead of {{lang|es|concepto}}, "concept") is not recommended except in {{lang|es|septiembre/setiembre}} ("September") and {{lang|es|séptimo}}/{{lang|es|sétimo}} ("seventh").<ref name="DPDp">{{cite book |title=Diccionario panhispánico de dudas |publisher=RAE-ASALE |edition=2nd (provisional) |url=https://www.rae.es/dpd/p |access-date=31 August 2023 |language=es |chapter=p}}</ref>
* An internal {{angbr|bs}} may be pronounced with a silent {{angbr|b}}. In some cases, this is considered vulgar; in others, it is the most common pronunciation and reflected in the orthography: {{lang|es|sustancia/substancia}} ("substance").<ref name="DPDb">{{cite book |title=Diccionario panhispánico de dudas |publisher=RAE-ASALE |edition=2nd (provisional) |url=https://www.rae.es/dpd/b |access-date=31 August 2023 |language=es |chapter=b}}</ref>
* Intervocalic or final {{angbr|d}} can be weakly pronounced or silent depending on formality and dialect, or in relaxed speech: {{lang|es-ES|[[Madrid]]}} {{IPAc-es|m|a|'|d|r|i|}}, {{lang|es-PR|cansado}} {{IPAc-es|k|a|n|'|s|a|o}} ("tired").<ref name="DPDd">{{cite book |title=Diccionario panhispánico de dudas |publisher=RAE-ASALE |edition=2nd (provisional) |url=https://www.rae.es/dpd/d |access-date=31 August 2023 |language=es |chapter=d}}</ref>

==Greek==

In [[Greek language|Greek]], the [[comma]] also functions as a silent letter in a handful of words, principally distinguishing {{lang|el|{{linktext|ό,τι}}}} (''ó,ti'', "whatever") from {{lang|el|{{linktext|ότι}}}} (''óti'', "that").<ref name=nicky>Nicolas, Nick. "[http://www.tlg.uci.edu/~opoudjis/unicode/punctuation.html Greek Unicode Issues: Punctuation] {{webarchive|url=https://web.archive.org/web/20150118083005/http://www.tlg.uci.edu/~opoudjis/unicode/punctuation.html |date=2015-01-18}}". 2005. Accessed 7 Oct 2014.</ref>

==Slavic languages==
{{Expand section|date=July 2018}}

===Czech===

{{Main|Czech phonology}}
In the vast majority of cases, Czech pronunciation follows the spelling. There are only four exceptions:

'''D'''

For example: dcera (daughter) and in srdce (heart)

====/j/ + consonant clusters in some words====

In most present forms of the verb ''být'' ("to be"), namely ''jsem'', ''jsi'', ''jsme'', ''jste'' and ''jsou'' (i.e., all persons but the third-person singular ''je''), the initial cluster /js/ is regularly simplified to a mere /s/. This pronunciation is considered correct and neutral when the verb is unstressed and used as an auxiliary. When stressed or used lexically, only the full /js/ pronunciation is considered correct. In casual speech, however, a few other highly frequent words commonly undergo similar simplification, namely all present forms of ''jít ''("to walk") beginning with /jd/ (that is ''jdu'', ''jdeš'', ''jde'', ''jdeme'', ''jdete'', ''jdou''), and the noun ''jméno'' ("name") (as well as the derived verb ''jmenovat (se)'' 'to name, to (''be'') call(''ed'')').<ref>Hejtmánková, J. (2017). Czech for English speakers (2nd ed.). Brno, Czech Republic: Edika. Page 34.</ref><ref>Janáček, L., & Cheek, T. (2017). The Janáček opera libretti: Translations and pronunciation. Page 43.</ref>

===Russian===

Several words in Russian omit written consonants when spoken. For example, "чувствовать" (chuvstvovat') is pronounced [ˈt͡ɕustvəvətʲ] and "солнце" (solntse) is pronounced [ˈsont͡sə].

Russian letter [[ъ]]  has no phonetic value and functions as a separation sign. Before the spelling reform of 1918, this hard sign was written at the end of each word when following a non-palatal consonant.

=== Ukrainian ===
Some three-consonant sequences in Ukrainian omit the second sound; for example, {{Lang|uk|шістнадцять}} (šistnadcjatj) is pronounced without the first t.

==Semitic languages==
[[File:00-Alif-Wasla.png|100px|thumbnail|right|The silent Arabic alif is marked with a [[Wasla (diacritic)|wasla]] sign above it ]]
In [[Hebrew language|Hebrew]], almost all cases of silent letters are silent [[aleph]] ({{Script/Hebr|א}}).<ref>A rare example for a Hebrew silent letter, which is not a silent aleph, is in the word {{lang|he|rtl=yes|יִשָּׂשכָר}} (meaning [[Issachar]]). In this word, the silent letter is equivalent to the English letter S. This word sounds like "ysachar", but is spelled like "ysaschar".</ref> Many words that have a silent aleph in Hebrew have an equivalent word in the [[Arabic language]], that is written with a [[mater lectionis]] [[Aleph|alif]] ({{script|Arab|[[ا]]}}), a letter that indicates the long vowel "aa". Examples:
* The Hebrew word for "no" is {{lang|he|rtl=yes|לֹא}} (sounds like "lo", spelled like "loa") and the Arabic word for "no" is {{lang|ar|rtl=yes|لاَ}} (sounds and spelled like "laa").
* The Hebrew word for "left side" is {{lang|he|rtl=yes|שְׂמֹאל}} (sounds like "smol", spelled like "smoal") and an Arabic word for "left side" is {{lang|ar|rtl=yes|شِمِال}} (sounds and spelled like "shimaal").
* The Hebrew word for "head" is {{lang|he|rtl=yes|רֹאשׁ}} (sounds like "rosh", spelled like "roash") and the Arabic word for "head" is {{lang|ar|rtl=yes|رَأس}} (sounds and spelled like "ra's").

The explanation for this phenomenon is that the Hebrew language had a [[sound change]] of all the mater lectionis aleph letters into silent ones (see [[Canaanite shift]]). Due to that sound change, in Hebrew language, there are only two kinds of aleph - the [[glottal stop]] (/ʔ/) and the silent one,<ref>{{Cite book|url=https://books.google.com/books?id=LhJuiY1Io84C&dq=glottal+stop+VS+silent+Aleph&pg=PA1|title=The Cambridge Biblical Hebrew Workbook: Introductory Level|first=Nava|last=Bergman|date=7 April 2005|publisher=Cambridge University Press|isbn=9780521826310 |via=Google Books}}</ref> while in Arabic language all three kinds still exist.<ref>{{Cite book|url=https://books.google.com/books?id=kRIHCnC74BoC&dq=glottal+stop+VS+silent+Alif&pg=PA32|title=Introduction to Arabic Natural Language Processing|first=Nizar Y.|last=Habash|date=15 March 2010|publisher=Morgan & Claypool Publishers|isbn=9781598297959 |via=Google Books}}</ref>

The silent Arabic alif is marked with a [[Wasla (diacritic)|wasla]] sign above it (see picture), in order to differentiate it from the other kinds of alifs. An Arabic alif turns silent, if it fulfills three conditions: it must be in a beginning of a word, the word must not be the first one of the sentence, and the word must belong to one of the following groups:
* [[Verb]]s that start with the [[prefix]] "ʔi-", due to their [[Grammatical conjugation|conjugation]] and [[derived stem]].
* Ten specific [[noun]]s that begin with "ʔ": اسم, است, ابن/ابنة, اثنان/اثنتان, امرؤ/امرأة, اَيمن الله/اَيْم الله. Some of these words have a Hebrew word equivalent, and that equivalent had totally lost the beginning aleph. Examples: {{lang|ar|rtl=yes|اسم}} (ʔism), meaning "a name" (in Maltese the word ''isem''), sounds like "ism" if it is in the beginning of the sentence and "sm" if not; its Hebrew equivalent is {{lang|he|rtl=yes|שֵׁם}} (shem). {{lang|ar|rtl=yes|إبن}} (ʔibn) (in Maltese, the word ''iben''), meaning "a son", sounds like "ibn" if it is in the beginning of the sentence and "bn" if not; its Hebrew equivalent is {{lang|he|rtl=yes|בֵּן}} (ben), in Maltese ''bin''.
* The alif of the word {{lang|ar|rtl=yes|اَل}} (ʔal), meaning "[[Definiteness|the]]", sounds like "al" if it is in the beginning of the sentence and "l" if not.

Besides the alif of the Arabic word {{lang|ar|rtl=yes|ال}} (ʔal, meaning "[[Definiteness|the]]"), its [[lām]] (the letter L) can also be silent. It becomes silent if the noun that word is related to starts with a "[[Sun and moon letters|sun letter]]". A sun letter is a letter that indicates a [[consonant]] produced by stopping the air in the front part of the mouth (not including the consonant M). The Hebrew equivalent to the Arabic word {{lang|ar|rtl=yes|ال}} (ʔal, meaning "[[Definiteness|the]]") had totally lost its L.

In Maltese, għ can be silent (e.g., ''għar'' 'cave', pronounced "ahr"), or /ħ/ if it is at the end of a word (e.g., ''qlugħ'' 'sail').<ref>{{Cite web |title=World Explorer |url=https://elearn.fiu.edu/e-dev/WorldExplorer/continents/europe/malta/malta_language.htm |archive-url=https://web.archive.org/web/20181219044355/https://elearn.fiu.edu/e-dev/WorldExplorer/continents/europe/malta/malta_language.htm |archive-date=19 December 2018 |website=[[Florida International University]]}}</ref>

==Turkish==

In the [[Turkish language]], {{angbr|[[ğ]]}} often has no sound of its own, but lengthens the preceding vowel, for example in ''dağ'' ("mountain") {{IPA|tr|daː|}}. In other surroundings, it may be pronounced as a [[semivowel|glide]].

== Persian ==
In [[Persian language|Persian]], there are two instances of silent letters:

* The letter [[He (letter)|he]] after a short vowel, unless in a monosyllabic word, has no sound of its own. It is only written because according to spelling rules a word cannot end in a short vowel.<ref name=":0">{{Cite web |date=2007 |title=Persian Online – Grammar & Resources » Silent and Consonantal /h/ |url=https://sites.la.utexas.edu/persian_online_resources/phonology/silent-letters-silent-and-consonantal/ |website=University of Texas at Austin College of Liberal Arts}}</ref>
* The [[Silent vav|Silent Vav]] is always written but not spoken in Standard Persian. It used to represent the [[labialization]] of the [[voiceless velar fricative]], which no longer exists in the standard dialect, making it an archaic remnant of the old standards of pronunciation.<ref name=":22">{{Cite web |date=2007 |title=Persian Online – Grammar & Resources » The Silent Letter vāv |url=https://sites.la.utexas.edu/persian_online_resources/phonology/silent-letters-vav/ |access-date=2022-07-23 |website=University of Texas at Austin College of Liberal Arts}}</ref>

==Indic languages==

Unconventional to [[Sanskrit]] and Proto-Indo-European root languages, some Indic languages have silent letters. Among [[Dravidian language]]s, [[Tamil language|Tamil]] and [[Malayalam]] have certain distinct styles of keeping few of their letters silent. Among the [[Indo-Aryan languages]], [[Bengali language|Bengali]] language has silent letters.

===Tamil===
{{unreferenced section|date=December 2018}}
[[Tamil language|Tamil]] is a classical language phonetically characterized by [[allophonic|allophones]], [[approximants]], [[nasal stop|nasals]] and [[Glottal stop|glottalised]] sounds. Some words, however, have silent letters in them. The words அஃது (while that is), and அஃதன் (that) contain the ''[[Āytam]]'' or '{{lang|ta|ஃ}}', which is not pronounced in Modern Tamil. It is explained in the ''Tolkāppiyam'' that ''āytam'' could have [[Glottal stop|glottalised]] the sounds it was combined with, though some may argue it sounded more like the Arabic '{{lang|ar|خ|rtl=yes}}' ({{IPA|/x/}}). That being said, modern words like ஆஃபிஸ் (Office) use '{{lang|ta|ஃ}}' and '{{lang|ta|ப}}' in sequence to represent the {{IPAc-en|f}} sound, as the ''āytam'' is nowadays also used to transcribe it and other foreign phonemes.

Another convention in [[Dravidian languages#Prehistory|Middle Tamil (Sen-Tamil)]] is the use of silent vowels to address a mark of respect when beginning proper nouns. The [[Ramayana]] was one such text where the word ''[[Ramayana]]'' in Tamil always began with '{{lang|ta|இ}}', as in [[Ramayana|இராமாயணம்]] ({{IPA|/ɾɑːmɑːjʌɳʌm/}}), though it was not pronounced. The name கோபாலன் ({{IPA|/ɡoːbɑːlʌɳ/}}) was so written as உகோபாலன் prefixed with an '{{lang|ta|உ}}'.

===Malayalam===
{{unreferenced section|date=December 2018}}

Inheriting [[elision]], [[approximant]]s and [[allophone]]s from [[Tamil language|Tamil]], in Malayalam, except for Sanskrit words, words ending in the vowel '{{lang|ml|ഉ}}' ({{IPA|/u/}}) become silent at the end and if not compounded with words succeeding them, replace the '{{lang|ml|ഉ}}' vowel by the schwa {{IPA|/ə/}}. However, it is considered disrespectful to change this pronunciation in the simple present verbs, when using imperatives and using what can be termed as Imperative-Active voice in Malayalam, where the second person is respectfully addressed with his or her name instead of നീ ({{IPA|/n̪i:/}}, you) or നിങ്ങൽ ({{IPA|/n̪iŋaɭ/}}, yourselves). For example, in the sentence, രാകേശ് പണി തീർക്കു ({{IPA|/ɾʲaːkeːɕə paɳi ti:ɾʲku/}}, Rakesh, finish your work), the use of the second personal pronoun is avoided with the name രാകേശ് ({{IPA|/ɾʲaːkeːɕ/}}, Rakesh), but this sentence sounds less respectful if the '{{lang|ml|ഉ}}' in തീർക്കു ({{IPA|/ti:ɾʲku/}}, finish} is replaced by the schwa or {{IPA|/ə/}}, as in "തീർക്കു!" ({{IPA|/ti:ɾʲkə/}}, Finish!) which sounds like an order. Notice the {{IPA|/ə/}} at the end of the name Rakesh which is pronounced after being added to the Sanskritic name.

===Bengali ===
Unlike other Indic languages, Bengali features silent consonants, which occur in many consonant clusters. These silent letters usually occur in loanwords borrowed from Sanskrit. These silent letters occur due to sound mergers as the spellings of Sanskrit loanwords have been preserved but their pronunciation has changed with sound mergers.

The letter ব ('b') is silent in most of the consonant clusters where it occurs as the second one. For example, স্বপ্ন (স্ব = স্ 'sh' + ব) (dream) is written as "'''shb'''ôpno" but pronounced as "'''sh'''ôpno".
জ্বর (জ্ব = জ্ 'j' + ব) (fever) is written as "'''jb'''ôr" but pronounced as "'''j'''ôr". This is the case with consonant clusters at the beginning of the words.
If the consonant cluster occurs in the middle or at the end of a word, the ব serves as a marker to put stress on the first consonant in a consonant cluster. For example, বিশ্বাস (শ্ব = শ্ 'sh' + ব) (to believe) is written as " bi'''shb'''āsh" but pronounced as " bi'''sh'''āsh" with more stress on the '''sh''' than usual, which sounds like "bi'''shsh'''āsh".

The letter ম ('m') also remains silent in many initial consonant clusters. For example, "'''স্মৃ'''তি" (স্মৃ = স্ 's' + ম + ঋ 'ri') (memory) is written as "'''smri'''ti" but pronounced as "'''sri'''ti". In many cases, if the consonant cluster occurs in the middle of a word, then the preceding vowel should be nasalised and the first letter in the cluster is stressed and ম in the cluster is silent. For example, আ'''ত্মা''' (ত্মা = ত্ 't' + মা 'mā'), i.e., "ā'''tmā'''" (soul), is pronounced as "''ā'''''ttā'''" and the ''ā'' is nasalised.

The letter য় ('y') is also silent in many cases as in "মে'''য়ে'''" (য়ে = য় + এ 'ē') (girl) is written as "mē'''yē'''" but pronounced as "mē".

The letter 'য' ('j') in its consonant clusters changes the pronunciation of the other letters in the cluster. For example, '''ন্যা'''য় (ন্যা = ন 'n'+ য + আ 'ā') (justice) is written as "'''njā'''y" but pronounced as "'''na'''y" ('''a''' as in h'''a'''t); ক'''ন্যা''' (girl or daughter) is written as "ko'''njā'''" but pronounced as "ko'''nnā'''". Sometimes it is completely silent as in স'''ন্ধ্যা''' (ন্ধ্যা = ন্ 'n' + ধ 'dh' + য + আ 'ā') (evening) is written as "sho'''ndhjā'''" but pronounced as "sho'''ndhā'''".

Moreover, Bengali also features schwa deletion common to other Indo-Aryan languages, where the schwa, 'o' or 'ô' is omitted while pronunciation, for example, কাকতলা (incident) is written as "kā'''ko'''tôlā" but pronounced as "kā'''k'''tôlā".

Similarly, in many other consonant clusters, the second consonant is silent.

== Zhuang-Tai languages ==

=== Thai ===
[[Thai language|Thai]] has a [[Orthographic depth|deep orthography]] like English and French. Unlike the two languages, however, the [[Thai script]] is an [[abugida]] rather than a true alphabet. Nonetheless, silent consonants, vowels, and even syllables are common in Thai. Thai has many loanwords from Sanskrit and [[Pali]], and rather than spell aforementioned words according to Thai phonics, the script tends to maintain the etymological spellings. For example, a romanization of the word ประโยชน์ that reflects Thai orthography is {{transliteration|th|prayochṅ}}, but it would be pronounced as {{transliteration|th|prayot}}, where the extra letter for ''-n'' is completely silent.<ref>{{Cite book|last=Juyaso|first=Arthit|title=Read Thai in 10 Days|publisher=Bing-Lingo|year=2016|isbn=978-616-423-487-1}}</ref> Another example is the Thai word มนตร์, which is sometimes written as ''mantra'' like it would be in Sanskrit, but it is only pronounced {{transliteration|th|mon}} in Thai. Though the second syllable is pronounced in Sanskrit, it is completely absent when pronouncing the word in Thai. In such words, the diacritic {{Lang|th|◌์}}, known as {{transliteration|th|thanthakhat}} ({{Langx|th|ทัณฑฆาต}}), is used to mark silent letters.

Also, different letters can be used for the same sound (for example, [tʰ] can be spelled as {{Lang|th|ฐ}}, {{Lang|th|ฑ}}, {{Lang|th|ฒ}}, {{Lang|th|ถ}}, {{Lang|th|ท}}, or {{Lang|th|ธ}}) depending on which class the consonant is, which is important for knowing which tone the syllable will have, and whether or not it is a loanword from Sanskrit or Pali. However, some letters written before low class consonants become silent and turn the low class syllable into a high class one. For example, even though the high class letter {{transliteration|th|ho hip}} {{Lang|th|ห}} is used to write the sound /h/, if the letter comes before a low class letter in a syllable, the letter will become {{transliteration|th|ho nam}}, which will make the letter silent and it will turn the syllable into a high class syllable. For example, the word {{Lang|th|นา}} is a low class syllable because its initial consonant is a low class consonant. The syllable is pronounced {{ipa|/nā:/}} (with a long vowel and mid tone) and it means "field". However, the word {{Lang|th|หนา}} is a high class syllable, despite it containing a low class consonant in the [[Onset (linguistics)|onset]]. The syllable is pronounced {{ipa|/nǎ:/}} (with a long vowel and a rising tone) and it means "thick".

=== Lao ===
Like Thai, [[Lao script|Lao]] also has a letter that becomes silent if it comes before a low class consonant. The letter is ''ho sung'' ຫ, which would represent the sound /h/ if it were not paired with another low class consonant. However, unlike Thai, the digraphs beginning with the aforementioned letter can sometimes be written as a [[Orthographic ligature|ligature]].

=== Zhuang ===

In the [[standard Zhuang]] language, written in the Latin script, the last letter of every syllable is typically silent due to it representing the tone of the syllable. The digraphs mb and nd also have silent letters, representing the phonemes ɓ and ɗ respectively.

== Korean ==
In the [[Hangul|Hangul Orthography]] of the [[Korean language]], the letter ⟨ㅇ⟩ is silent when written in the syllable-initial position, and represents the sound /ŋ/ when written in the syllable-final position. For example, in the word 안녕 ([[Yale romanization of Korean|Yale Romanization]]: {{transliteration|ko|annyeng}}) (meaning "hello"), composed of the letters "ㅇㅏㄴㄴㅕㅇ", the first ⟨ㅇ⟩ is silent, and the last ⟨ㅇ⟩ is pronounced as /ŋ/. The reason for this can be found in 15th-century Hangul orthography. In the 15th century, the letter ⟨ㅇ⟩ originally represented /∅~ɣ/ (a [[Lenition|lenited]] form of ㄱ /k/), while the letter ⟨ㆁ⟩ unconditionally represented /ŋ/. But because in Middle Korean phonology, ⟨ㆁ⟩ was not allowed in syllable-initial position, and ⟨ㅇ⟩ was not allowed in syllable-final position, it formed a complementary distribution of the two letters. Because of this and due to the fact that the letters look very much alike, the two letters merged.<ref>{{cite book |last1=Yi |first1=Ki-mun |last2=Ramsey |first2=S. Robert |title=A history of the Korean language |date=2011 |publisher=Cambridge University Press |location=Cambridge |isbn=978-0-521-66189-8}}</ref>

Korean's syllable structure is CGVC, and Korean's writing system, [[Hangul]], reflects this structure. The only possible consonant cluster in a single syllable must contain a [[Semivowel|glide]] and they must occur in the [[Onset (linguistics)|onset]]. However, sometimes a cluster of two consonants are written after the vowel in a syllable. In such situations, if the next syllable begins with a vowel sound, then the second consonant becomes the first sound of the next syllable. However, if the next syllable begins with a consonant sound, then one of the consonants in the cluster will be silent (sometimes causing [[fortition]] in the following consonant). For example, the word 얇다 (meaning "thin") is written as ([[Yale romanization of Korean|Yale]]: ''yalp.ta''), but the word is pronounced as if it was written ''yal.tta'' because the second syllable begins with a consonant sound. However, the word 얇아서 (also meaning "thin") is written as ([[Yale romanization of Korean|Yale]]: ''yalp.a.se'') and it is pronounced as ''yal.pa.se'' because the second syllable begins with a vowel sound.<ref>{{Cite web|url=https://freshkorean.com/tag/final-consonant/|title=Final Consonant|website=Fresh Korean|date=21 May 2013 |access-date=2020-04-15}}</ref>

== Mongolian ==
Interestingly, the native [[Mongolian script]] has much more [[orthographic depth]] than [[Mongolian Cyrillic alphabet|Mongolian Cyrillic]]. For example, the letter ''Gh'' or ''γ'' (ᠭ) is silent if it is between two of the same vowel letters. In that case, the silent consonant letter combines to two written vowel into one long vowel. For example, the Mongolian word ''Qaγan'' (ᠬᠠᠭᠠᠨ) should be pronounced ''Qaan'' (ᠬᠠᠠᠨ). In Mongolian Cyrillic, however, it is spelled ''хаан'' (''haan''), closer to the actual pronunciation of the word. Words in the Mongolian script can also have silent vowels as well. For Mongolian name of the city [[Hohhot]], it is spelled ''Kökeqota'' (ᠬᠥᠬᠡᠬᠣᠲᠠ) in Mongolian script, but in Cyrillic, it is spelled ''Хөх хот (Höh hot)'', closer towards the actual pronunciation of the word.

==Basque==
In Basque, during the 20th century {{angbr|h}} was not used in the orthography of the Basque dialects in Spain but it marked an aspiration in the North-Eastern dialects. During the standardization of Basque in the 1970s, the compromise was reached that {{angbr|h}} would be accepted if it were the first consonant in a syllable. Hence, {{lang|eu|herri}} ("people") and {{lang|eu|etorri}} ("to come") were accepted instead of {{lang|eu-biscayan|erri}} ([[Biscayan]]) and {{lang|eu-FR|ethorri}} ([[Souletin]]). Speakers could pronounce the h or not. For the dialects lacking the aspiration, this meant a complication added to the standardized spelling.

==See also==
*[[Apheresis (linguistics)]]
*[[Elision]]
*[[Schwa deletion in Indo-Aryan languages]], explains rules of Modern Indo-Aryan languages that delete the schwa sound.
*[[Silent e]]
*[[Silent k]]
*[[Syncope (phonology)]]
*[[Three-letter rule]], the source of some common English silent letters.

==References==
{{Reflist}}

[[Category:Silent letters| ]]
[[Category:Spelling]]
[[Category:English orthography]]
[[Category:Silence|Letter]]