phonotactics (phonology)

Glossary of Grammatical and Rhetorical Terms

Kerrching (or kerching or kaching) is an onomatopoeic word that mimics the sound of an old-fashioned cash register. In English, the consonant cluster /ng/ [ŋ] may appear at the end of a word but not (ordinarily) at the beginning of a word. (Jacquie Boyd/Getty Images)


In phonology, phonotactics is the study of the ways in which phonemes are allowed to combine in a particular language. (A phoneme is the smallest unit of sound capable of conveying a distinct meaning.) Adjective: phonotactic.

Over time, a language may undergo phonotactic variation and change. For example, as Daniel Schreier points out, "Old English phonotactics admitted a variety of consonantal sequences that are no longer found in contemporary varieties" (Consonant Change in English Worldwide, 2005).

Phonotactic constraints are rules and restrictions concerning the ways in which syllables can be created in a language. Linguist Elizabeth Zsiga observes that languages "do not allow random sequences of sounds; rather, the sound sequences a language allows are a systematic and predictable part of its structure."

Phonotactic constraints, says Zsiga, are "restrictions on the types of sounds that are allowed to occur next to each other or in particular positions in the word" ("The Sounds of Language" in An Introduction to Language and Linguistics, 2014).

According to Archibald A. Hill, the term phonotactics (from the Greek for "sound" + "arrange") was coined in 1954 by American linguist Robert P. Stockwell, who used the term in an unpublished lecture delivered at the Linguistic Institute in Georgetown.

See Examples and Observations below. Also see:

Examples and Observations

  • "Becoming sensitive to phonotactics is not only important for learning how sounds occur together; it is also crucial for discovering word boundaries."
    (Kyra Karmiloff and Annette Karmiloff-Smith, Pathways to Language. Harvard University Press, 2001)

  • Phonotactic Constraints in English
    - "Phonotactic constraints determine the syllable structure of a language. . . . Some languages (e.g. English) allow consonant clusters, others (e.g. Maori) do not. English consonant clusters are themselves subject to a number of phonotactic constraints. There are constraints in terms of length (four is the maximum number of consonants in a cluster, as in twelfths /twεlfθs/); there are also constraints in terms of what sequences are possible, and where in the syllable they can occur. For example, although /bl/ is a permissible sequence at the start of a syllable, it cannot occur at the end of one; conversely, /nk/ is permitted at the end, but not the start."
    (Michael Pearce, The Routledge Dictionary of English Language Studies. Routledge, 2007)

    - "She held her eyes open every minute, forgetting how to blink or nap."
    (Cynthia Ozick, "The Shawl." The New Yorker, 1981)

    - "Certain phonotactic constraints—that is, constraints on syllable structure—are thought to be universal: all languages have syllables with vowels, and all languages have syllables that consist of a consonant followed by a vowel. But there is also a great deal of language specificity in phonotactic constraints. A language like English allows just about any type of consonant to appear in the coda (syllable-final) position—try it yourself, by coming up with as many words as you can that add only one consonant to the sequence /k?_/, like kit. You will find there are many. In contrast, languages like Spanish and Japanese have strict constraints about syllable-final consonants."
    (Eva M. Fernández and Helen Smith Cairns, Fundamentals of Psycholinguistics. Wiley, 2011)

  • Arbitrary Phonotactic Constraints
    "Many of the phonotactic limitations are arbitrary, . . . not involving articulation, but depending only on the idiosyncrasies of the language in question. For example, English has a constraint that forbids the sequence of a stop followed by a nasal word initially; the sign # marks a boundary, a word boundary in this case, and the star means that what follows is ungrammatical:
    (28) Phonotactic constraint  Phonemic level: *#[+stop][+nasal]
    Thus, English words like knife and knee are pronounced /naɪf/ and /ni/. Historically, they did have the initial /k/, which is still present in several sister languages. . . . Phonotactic restrictions are thus not necessarily due to any articulatory difficulty, because what cannot be said in one language can be said in another. Rather, these constraints are very often due to changes taking place in one language, but not in the others, as the English, Swedish, and German cognates . . . demonstrate. The result of this historical change in English has created a discrepancy between orthography and pronunciation, but this discrepancy is not due to the change per se, but to the fact that the English orthography has not been revised. Should we want to keep up with today's pronunciation, knife and knee might be spelled 'nife' and 'nee,' ignoring, of course, the optimal spelling of the vowels."
    (Riitta Välimaa-Blum, Cognitive Phonology in Construction Grammar: Analytic Tools for Students of English. Walter de Gruyter, 2005)