What Is Phonotactics?

What does phonotactics mean?

Phonotactics is the part of phonology that studies which sequences of speech sounds are allowed in a language and where those sounds can appear. It helps explain why a sequence may sound perfectly ordinary to speakers of one language but unusual to speakers of another.[1]

Think about the beginning of the English word street. Its first three sounds, /s/, /t/, and /r/, occur together in many English words. Now imagine a word beginning with /bn/. Both /b/ and /n/ are familiar English sounds, but the combination is not a normal word-initial cluster in native English vocabulary.

The difference is not that English speakers are physically unable to pronounce /bn/. Rather, English does not usually organize the beginning of a word that way. Every language has its own patterns for arranging sounds, and speakers learn those patterns as they learn the language.

The central idea

Knowing which sounds a language uses is only part of knowing its sound system. You also need to know which sounds can go together, in what order, and in which positions.[2]

Phonotactics and syllable structure

Many phonotactic restrictions concern the syllable, a unit of sound built around a central, relatively sonorous element. In most syllables, that element is a vowel. Linguists commonly divide a syllable into an onset, a nucleus, and a coda.[1]

Take the one-syllable English word plant, pronounced /plænt/:

/pl/OnsetBefore the vowel
/æ/NucleusSyllable center
/nt/CodaAfter the vowel

The nucleus and coda together are called the rime (also spelled rhyme).

Not every syllable needs an onset or a coda. For example, me has an onset but no coda, while at starts with a vowel and has a coda. Syllables ending in a vowel are called open syllables; those ending in a consonant are called closed syllables.

Linguists often describe these shapes with C for consonant and V for vowel:

PatternExampleWhat it contains
CVme /miː/One consonant followed by a vowel
CVCcat /kæt/One consonant, a vowel, and a final consonant
CCVCCplant /plænt/Two consonants before and after the vowel
CCCVCspring /sprɪŋ/Three consonants at the beginning

English allows several different syllable shapes, including some with complex groups of consonants. Other languages permit fewer possibilities. Phonotactics describes both these broad syllable patterns and the more specific sound combinations allowed within them.[1]

Examples of phonotactic rules in English

English has many consonant clusters, but it does not allow consonants to be combined in just any order. A cluster is a group of adjacent consonant sounds within a syllable, such as the /pl/ of play or the /str/ of street.

Rules about sound combinations

At the beginning of ordinary native English words, clusters such as /pl/, /br/, and /str/ are possible. Clusters such as /tl/ and /dl/ are not normally allowed in the onset, even though both individual sounds occur in English. Other languages, including some that have words beginning with /tl/, organize these sounds differently.[2]

Sound sequenceEnglish exampleStatus
/pl/play, plantAllowed onset
/str/street, strongAllowed onset
/ŋ/sing (at the end)Not word-initial
/kn/knee has /n/, not /kn/Not a modern onset

Rules about sound positions

Phonotactics also determines where a sound can occur. The English sound /ŋ/, heard at the end of sing, is normal in that position, but it does not normally begin an English word. This restriction concerns the sound, not the letter used to write it.[1]

That distinction matters in English because spelling can hide changes in pronunciation. The written kn at the beginning of knee reflects an older pronunciation pattern. In Modern English, its /k/ is silent, so the word starts with /n/.

How phonotactics differs across languages

A sound pattern that is ordinary in one language may be restricted in another. These differences are not signs that one language is simpler or more advanced. They are simply different ways of organizing speech sounds.

EnglishsprayThe word starts directly with the consonant cluster /spr/.
SpanishesprayThe borrowed word adds /e/ before /spr/ because Spanish does not ordinarily start words with /s/ plus another consonant.

In Spanish, words generally do not begin with /s/ followed immediately by another consonant. This helps explain forms such as escuela (school), estudiante (student), and adapted borrowings such as espray (spray). English allows initial /sk/, /st/, and /spr/ clusters that Spanish typically avoids.[3]

German offers a different comparison. It permits a word-initial /kn/ cluster, as in Knoten (‘knot’). Modern English no longer uses that pronounced cluster in ordinary words, although spellings such as knot preserve evidence of it.[1]

Languages can also differ in how their syllables end. Hawaiian, for instance, has a particularly restrictive native syllable pattern: syllables end in vowels rather than consonants. English can end syllables with consonants and even complex clusters, as in texts or plants.[1]

Why some made-up words sound believable

Not every possible sound sequence is an actual word. This creates an important difference between a word that could belong to a language and a sequence that does not fit its normal sound rules.

Consider the invented word stremp. It is not a standard English word, but its beginning /str/ and ending /mp/ are familiar English patterns. It therefore sounds like something that could be a word. Linguists call an unoccupied but permissible possibility an accidental gap.[1]

Now compare it with a made-up form pronounced /bnɛmp/. Because /bn/ is not a normal initial English cluster, this sequence violates the familiar pattern. An invented form of this kind is phonotactically ill-formed in ordinary English.

Two kinds of nonwords

Possible but unused: stremp resembles ordinary English sound patterns.
Outside the usual pattern: /bnɛmp/ starts with a cluster that ordinary English does not permit.

There is also a difference between what is permitted and what is common. Some legal sound combinations occur much more frequently than others. Researchers call this phonotactic probability, and it helps explain why some unfamiliar sound sequences feel more word-like than other equally permissible ones.[5]

The role of sonority

One reason languages favor certain consonant orders involves sonority, a term for how prominent or vowel-like a sound is. Vowels are generally highly sonorous, while stops such as /p/ and /t/ are much less sonorous.

A common tendency is for sonority to rise toward the vowel at the center of a syllable and fall afterward. The English word play illustrates this: the onset goes from the stop /p/ to the more sonorous /l/, then to the vowel. Reversing the consonants to create /lp/ would be unusual as an English syllable onset.[1]

This tendency is often called the sonority sequencing principle, but it is not a rule without exceptions. English words such as spin begin with /sp/, where sonority does not rise smoothly. Individual languages have specific phonotactic patterns that cannot be explained by sonority alone.

What happens when languages borrow words?

Borrowing a word sometimes means borrowing a sound sequence that does not fit neatly into the receiving language. Speakers may adapt the word to make it more compatible with familiar pronunciation patterns.

One common strategy is epenthesis, the insertion of a sound. Japanese, for example, often introduces vowels when adapting English words with consonant clusters or certain final consonants. Research on Japanese loanwords shows that both vowel insertion and consonant deletion can occur, depending partly on the sounds and how the borrowed word is perceived.[4]

Spanish offers another kind of adaptation: an initial vowel can appear before a borrowed word beginning with /s/ plus a consonant. This process helps explain why English spray has the Spanish adaptation espray.[3]

Not all borrowings change in the same way. Some speakers and communities retain unusual foreign clusters, especially in names or well-established loanwords. Phonotactic patterns can also change over time as languages come into contact with one another.

Why phonotactics matters

Phonotactics is more than a list of allowed and disallowed clusters. It is part of the sound knowledge speakers use to recognize, learn, and produce words.

  • Learning pronunciation: A learner may add a vowel to a difficult cluster or leave out a sound when the target sequence is unfamiliar in their first language.
  • Recognizing spoken words: Familiar sound patterns can make an unfamiliar sequence easier to process. Research shows that the frequency of sound combinations affects how word-like a nonword sounds.[5]
  • Understanding language change: Shifts in pronunciation and the adaptation of borrowed words show how sound patterns develop over time.
  • Studying new words: Researchers can create nonwords that follow a language’s sound patterns to investigate pronunciation, perception, and vocabulary learning.

These patterns usually operate without conscious effort. Most English speakers do not need to study phonology to recognize that stremp sounds more like a possible English word than /bnɛmp/. Their experience with English has already taught them which sound sequences to expect.

Sources

  • [1]Cox, F., Harrington, J., and Mannell, R. “Syllable and Foot.” Macquarie University, Department of Linguistics. Read the sourceSyllable structure, English phonotactics, sonority, and accidental gaps
  • [2]“Phonotactics and Natural Classes.” Essentials of Linguistics, 2nd edition, eCampusOntario. Read the sourceLanguage-specific restrictions and English onset clusters
  • [3]“Teaching Spanish Pronunciation.” The Open University, OpenLearn. Read the sourceDifferences between English and Spanish phonotactics
  • [4]Shoji, S., and Shoji, K. “Vowel Epenthesis and Consonant Deletion in Japanese Loanwords from English.” Proceedings of the Annual Meetings on Phonology (2014). Read the sourceResearch on sound adaptation in borrowed words
  • [5]“Phonotactic Probability.” Spoken Language Laboratory, University of Kansas. Read the sourceFrequency of sound combinations and word processing

Related articles