allophone (word sounds)

Glossary of Grammatical and Rhetorical Terms

Examples of five allophones of the phoneme /t/ in General American English.

In linguistics, an allophone is an audibly distinct variant of a phoneme, such as the different pronunciations of the t sound in tar and star. Adjective: allophonic.

Substituting one allophone for another allophone of the same phoneme doesn't lead to a different word, just a different pronunciation of the same word. For this reason, allophones are said to be noncontrastive.

From the Greek, "other" + "sound"

Examples and Observations

  • "[E]very speech sound we utter is an allophone of some phoneme and can be grouped together with other phonetically similar sounds." (William O'Grady, et al., Contemporary Linguistics. Bedford, 2001)
  • "The allophones of a phoneme form a set of sounds that (1) do not change the meaning of a word, (2) are all very similar to one another, and (3) occur in phonetic contexts different from one another—for example, syllable initial as opposed to syllable final. The differences between allophones can be stated in terms of phonological rules." (Peter Ladefoged and Keith Johnson, A Course in Phonetics, 6th ed. Wadsworth, 2011)

Different Relationships, Different Sounds

"Many allophones, that is actual articulations, are possible for any phoneme of a language, depending on individual people's pronunciation, but the main allophones of any particular language are conditioned by their relationship to the surrounding sounds.

Thus in standard English the /l/ phoneme has a clear sound when it precedes a vowel (as in listen or fall in); a somewhat devoiced sound when preceded by a voiceless plosive (as in please, clue), and a dark sound when it occurs word-finally after a vowel (as in fall down) or when it is syllabic (as in muddle)." (Sylvia Chalker and Edmund Weiner, Oxford Dictionary of English Grammar.

Oxford University Press, 1998)

Allophones vs. Phonemes

"Sounds that are merely phonetic variants of the same phoneme are allophones. Notice that any two sounds of a given language represent either two allophones of the same phoneme (if the sounds can be interchanged in words with no resulting change in meaning, such as the p's of pit and keep) or two different phonemes (if the sounds cannot be interchanged without a resulting change in meaning, such as the m and s of milk and silk). . . .

"Now consider the word stop. If you say the word several times, you will probably notice that sometimes the final /p/ contains more aspiration and sometimes, less. (In fact, if you end the word with your lips together and do not release the /p/, it contains no aspiration at all.) Since you are not pronouncing stop as part of a larger chunk of language that varies from utterance to utterance (for example, John told Mary to stop the car versus Stop and go versus When you come to the sign, stop), the phonetic environment of the /p/ remains constant—it is at the end of the word and preceded by /a/. In other words, we cannot predict when a particular allophone with more or less aspiration is likely to occur, so the allophones of /p/ must be in free variation." (Thomas Murray, The Structure of English.

Allyn and Bacon, 1995)

An Infinite Number of Allophones

"[T]he choice of one allophone rather than another may depend on such factors as communicative situation, language variety, and social class. . . . [W]hen we consider the wide range of possible realisations of any given phoneme (even by a single speaker), it becomes clear that we owe the vast majority of allophones in free variation to idiolects or simply to chance, and that the number of such allophones is virtually infinite." (Paul Skandera and Peter Burleigh, A Manual of English Phonetics and Phonology. Gunter Narr Verlag, 2005)

Pronunciation: AL-eh-fon

mla apa chicago
Your Citation
Nordquist, Richard. "allophone (word sounds)." ThoughtCo, Apr. 25, 2017, Nordquist, Richard. (2017, April 25). allophone (word sounds). Retrieved from Nordquist, Richard. "allophone (word sounds)." ThoughtCo. (accessed January 18, 2018).