Silent letter

In an alphabetic writing system, a silent letter is a letter that, in a particular word, does not correspond to any sound in the word's pronunciation. Silent letters create problems for both native and non-native speakers of a language, as they make it more difficult to guess the spellings of spoken words or the pronunciations of written words. Newly developed alphabets for previously unwritten languages are thus typically designed to have no silent letters. Likewise, planned languages such as Interlingua and Esperanto tend to avoid silent letters.

Phonetic transcriptions that better depict pronunciation and which note changes due to grammar and proximity of other words require a symbol to show that the letter is mute. Handwritten notes use a circle with a line through it and the sound is called "zero"; it resembles the symbol for the "empty set", but must not be confused with the Danish letter Ø. In printed or computer graphic presentation using the IPA system, the symbol ∅ is used, which is like a diamond with a slash through it.

English

One of the noted difficulties of English spelling is a high number of silent letters. Carney distinguishes different kinds of "silent" letter, which present differing degrees of difficulty to readers and writers.

• Auxiliary letters which constitute digraphs, where two letters combine to 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". There are examples
    • where the phoneme has no standard single-letter representation, as with consonants <ng> for /ŋ/ as in sing, <th> for /θ/ as in thin or /ð/ as in then, and <sh> for /ʃ/ as in show, and diphthongs <ou> in out or <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 <gh> in enough or <ph> in physical instead of <f>. These are irregular for writers but may be 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 doubled consonants, as <bb> in clubbed; though not geminate consonants, as <ss> in misspell. Doubling due to suffixation or inflection is regular; otherwise it may present difficulty to writers (e.g. accommodate is often misspelt) but not to readers.
    • the discontiguous digraphs whose second element is "magic e", e.g. <a_e> in rate, <i_e> in fine. This is the regular way to represent "long" vowels in the last syllable of a morpheme.
    • others such as <ck> (which is in effect the "doubled" form of <c>), <gu> as in guard, vogue; <ea> as in bread, heavy, etc. These are difficult for writers and sometimes for readers.
• Dummy letters which bear no relation to neighbouring letters and have no correspondence in pronunciation
  • Some are inert letters, where the letter is sounded in a cognate word: e.g. <n> in damn (cf. damnation); <g> in phlegm (cf. phlegmatic); <a> in practically (cf. practical). 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. <w> in answer, <h> in honest, <s> in island, <b> in subtle. These present the greatest difficulty to writers and often to readers.

The distinction between "endocentric" digraphs and empty letters is somewhat arbitrary. For example, in such words as little and bottle one might view <le> as an "endocentric" digraph for /l̩/, or view <e> as an empty letter; similarly with <bu> or <u> in buy and build.

Not all silent letters are completely redundant:

• Silent letters can distinguish between homophones, 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 or origin of a word, e.g. vineyard suggests vines more than the phonetic *vinyard would.
• The final <fe> in giraffe gives a clue to the second-syllable stress, where *giraf might suggest initial-stress.

Silent letters arise in several ways:

• Pronunciation changes occurring without a spelling change. The <gh> spelling was in Old English pronounced [x] in such words as light.
• Sound distinctions from foreign languages may be lost, as with the distinction between smooth rho (ρ) and roughly aspirated rho (ῥ) in Ancient Greek, represented by <r> and <rh> in Latin, but merged to the same [r] in English. Similarly with <f> / <ph>, the latter from Greek phi.
• Clusters of consonants may be simplified, producing silent letters e.g. silent <th> in asthma, silent <t> in Christmas. Similarly with alien clusters such as Greek initial <ps> in psychology and <mn> in mnemonic.
• Occasionally, spurious letters are consciously inserted in spelling. The <b> in debt and doubt was inserted to reflect Latin cognates like debit and dubitable.

Since accent and pronunciation differ, letters may be silent for some speakers but not others. In non-rhotic accents, <r> is silent in such words as hard, feathered; in h-dropping accents, <h> is silent. A speaker may pronounce <t> in "often" or "tsunami" or neither or both.

See also

• Silent E
• Silent k
• Three letter rule source of some common English silent letters.
• List of names in English with non-intuitive pronunciations, many with multiple silent letters.

References

• Carney, Edward (1994). A Survey of English Spelling. Routledge. ISBN 0-415-09270-1. §2.6.5