View source for Phonotactics

{{Short description|Sounds allowed in a language (phonetics)}}
{{more footnotes needed|date=April 2009}}
{{expert needed|linguistics|reason=More details are needed on English phonotactics; phonotactics for other languages need to be discussed; further needs to be said about universals or the lack thereof; see [[Talk:Phonotactics#for expansion|the talk page]] for more possible expansions|date=March 2011}}
{{IPA notice}}
'''Phonotactics''' (from [[Ancient Greek]] {{transliteration|grc|phōnḗ}} {{gloss|voice, sound}} and {{transliteration|grc|taktikós}} {{gloss|having to do with arranging}})<ref>{{LSJ|fwnh/|φωνή}}, {{LSJ|taktiko/s|τακτικός|ref}}</ref> is a branch of [[phonology]] that deals with restrictions in a [[language]] on the permissible combinations of [[phoneme]]s. Phonotactics defines permissible [[syllable]] structure, [[consonant cluster]]s and [[vowel]] sequences by means of ''phonotactic constraints''.

Phonotactic constraints are highly language-specific. For example, in [[Japanese language|Japanese]], consonant clusters like {{IPA|/rv/}} do not occur. Similarly, the clusters {{IPA|/kn/}} and {{IPA|/ɡn/}} are not permitted at the beginning of a word in Modern English but are permitted in [[German language|German]] and were permitted in [[Old English|Old]] and [[Middle English]].<ref>{{Cite journal |last1=Orzechowska |first1=Paula |last2=Wiese |first2=Richard |date=2015-01-01 |title=Preferences and variation in word-initial phonotactics: A multi-dimensional evaluation of German and Polish |url=http://dx.doi.org/10.1515/flin-2015-0016 |journal=Folia Linguistica |volume=49 |issue=2 |doi=10.1515/flin-2015-0016 |issn=0165-4004|url-access=subscription }}</ref> In contrast, in some [[Slavic languages]] {{IPA|/l/}} and {{IPA|/r/}} are used alongside vowels as syllable nuclei.

Syllables have the following internal segmental structure:
* [[Syllable onset|Onset]] (optional)
* [[Syllable rhyme|Rhyme]] (obligatory, comprises nucleus and coda):
** [[Syllable nucleus|Nucleus]] (obligatory)
** [[Syllable coda|Coda]] (optional)

Both onset and coda may be empty, forming a vowel-only syllable, or alternatively, the nucleus can be occupied by a [[syllabic consonant]]. Phonotactics is known to affect [[second language]] [[vocabulary acquisition]].{{sfn|Laufer|1997}}

==Sonority sequencing principle==
Segments of a syllable are universally distributed following the [[sonority sequencing principle]] (SSP), which states that, in any syllable, the nucleus has maximal sonority and that sonority decreases as you move away from the nucleus. Sonority is a measure of the amplitude of a speech sound. The particular ranking of each speech sound by sonority, called the [[sonority hierarchy]], is language-specific, but, in its broad lines, hardly varies from a language to another,<ref>{{Cite journal|last1=Jany|first1=Carmen|last2=Gordon|first2=Matthew|last3=Nash|first3=Carlos M|last4=Takara|first4=Nobutaka|date=2007-01-01|title=HOW UNIVERSAL IS THE SONORITY HIERARCHY?: A CROSS-LINGUISTIC ACOUSTIC STUDY|url=https://www.researchgate.net/publication/242234684|journal=ICPhS|volume=16|pages=1096}}</ref> which means all languages form their syllables in approximately the same way with regards to sonority.

To illustrate the SSP, the [[voiceless alveolar fricative]] {{IPA|[s]}} is lower on the sonority hierarchy than the [[alveolar lateral approximant]] {{IPA|[l]}}, so the combination {{IPA|/sl/}} is permitted in onsets and {{IPA|/ls/}} is permitted in codas, but {{IPA|/ls/}} is not allowed in onsets and {{IPA|/sl/}} is not allowed in codas. Hence ''slips'' {{IPA|/slɪps/}} and ''pulse'' {{IPA|/pʌls/}} are possible English words while ''*lsips'' and ''*pusl'' are not.

The SSP expresses a very strong cross-linguistic tendency, however, it does not account for the patterns of all complex syllable margins, as there are both initial as well as final clusters violation the SSP, in two ways: the first occurs when two segments in a margin have the same sonority, which is known as a ''sonority plateau''. Such margins are found in a few languages, including English, as in the words ''sphinx'' and ''fact'' (though note that ''phsinx'' and ''fatc'' both violate English phonotactics).

The second instance of violation of the SSP is when a peripheral segment of a margin has a higher sonority than a segment closer to the nucleus. These margins are known as reversals and occur in some languages including English (''steal''  {{IPA|[stiːɫ]}}, ''bets''  {{IPA|/bɛts/}}) or French (''{{lang|fr|dextre}}'' {{IPA|/dɛkstʁ/}} but originally {{IPA|/dɛkstʁə/}}, ''{{lang|fr|strict}}'' {{IPA|/stʁikt/}}).<ref>{{Cite journal|last=Carlisle|first=Robert S.|date=2001-06-01|title=Syllable structure universals and second language acquisition|url=https://revistas.um.es/ijes/article/view/47581|journal=International Journal of English Studies|volume=1|issue=1|issn=1578-7044}}</ref>

==Examples==
===English===
{{main|English phonology#Phonotactics}}

The English syllable (and word) ''twelfths'' {{IPA|/twɛlfθs/}} is divided into the onset {{IPA|/tw/}}, the nucleus {{IPA|/ɛ/}} and the coda {{IPA|/lfθs/}}; thus, it can be described as CCVCCCC (C = consonant, V = vowel). On this basis it is possible to form rules for which representations of phoneme classes may fill the cluster. For instance, English allows at most three consonants in an onset, but among native words under standard accents (and excluding a few obscure loanwords such as ''[[sphragistics]]''), phonemes in a three-consonantal onset are limited to the following scheme:<ref>{{cite book|last=Crystal |first=David |author-link=David Crystal |title=The Cambridge Encyclopedia of the English Language |publisher=[[Cambridge University Press]] |year=2003 |isbn=978-0-521-53033-0 |page=243}}</ref>

:{{IPA|/s/}} + [[stop consonant|stop]] + [[approximant consonant|approximant]]:
:*{{IPA|/s/}} + {{IPA|/t/}} + {{IPA|/r/}}
::*''strap''
:*{{IPA|/s/}} + {{IPA|/t/}} + {{IPA|/j/}} (not in most accents of [[American English]])
::*''stew''
:*{{IPA|/s/}} + {{IPA|/p/}} + {{IPA|/j r l/}}
::*''sputum''
::*''sprawl''
::*''splat''
:*{{IPA|/s/}} + {{IPA|/k/}} + {{IPA|/j r l w/}}
::*''skew''
::*''scream''
::*''sclerosis''
::*''squirrel''

This constraint can be observed in the pronunciation of the word ''blue'': originally, the vowel of ''bl'''ue''''' was identical to the vowel of ''c'''ue''''', approximately {{IPA|[iw]}}. In most dialects of English, {{IPA|[iw]}} shifted to {{IPA|[juː]}}. Theoretically, this would produce {{IPA|*[bljuː]}}. The cluster {{IPA|[blj]}}, however, infringes the constraint for three-consonantal onsets in English. Therefore, the pronunciation has been reduced to {{IPA|[bluː]}} by [[elision]] of the {{IPA|[j]}} in what is known as [[yod-dropping]].

Not all languages have this constraint; compare [[Spanish language|Spanish]] {{lang|es|'''pli'''egue}} {{IPA|[ˈpljeɣe]}} or [[French language|French]] {{lang|fr|'''plu'''ie}} {{IPA|[plɥi]}}.

Constraints on English phonotactics include:<ref>{{cite book|last1=Harley |first1=Heidi |author-link1=Heidi Harley |title=English Words: A Linguistic Introduction |publisher=[[Wiley-Blackwell]] |year=2003 |isbn=0631230327 |pages=58–69}}</ref>

* All syllables have a [[Syllable nucleus|nucleus]]
* No [[geminate consonant]]s
* No onset {{IPA|/ŋ/}}
* No {{IPA|/h/}} in the [[syllable coda]] (except in [[Hiberno-English]])
* No [[affricate]]s in complex onsets (except when underlying {{IPA|/tr/}} and {{IPA|/dr/}} are analysed as {{IPA|/tʃr/}} and {{IPA|/dʒr/}})<ref>{{cite journal |last1=Schwartz |first1=Geoffrey |title=All TRs are not created equal: L1 and L2 perception of English cluster affrication |journal=Journal of Linguistics |publisher=Cambridge University Press |pages=623–654 |language=en |doi=10.1017/S0022226722000275 |date=August 2023|volume=59 |issue=3 |doi-access=free }}</ref>
* No {{IPA|/h/}} in complex onsets
* The first consonant in a complex onset must be an [[obstruent]] (e.g. '''''st'''op''; combinations such as ''*'''nt'''at'' or *'''''rk'''oop'', with a [[sonorant]], are not allowed)
* The second consonant in a complex onset must not be a voiced obstruent (e.g. ''*'''zd'''op'' does not occur)
* If the first consonant in a complex onset is not {{IPA|/s/}}, the second must be a [[Liquid consonant|liquid]] or a [[Semivowel|glide]]
* Every subsequence contained within a sequence of consonants must obey all the relevant phonotactic rules (the substring principle rule)
* No glides in syllable codas (excluding the {{linktext|offglide}}s of [[diphthong]]s)
* The second consonant in a complex coda must not be {{IPA|/r/}}, {{IPA|/ŋ/}}, {{IPA|/ʒ/}}, or {{IPA|/ð/}} (compare ''[[asthma]]'', typically pronounced {{IPAc-en|ˈ|æ|z|m|ə}} or {{IPAc-en|ˈ|æ|s|m|ə}}, but rarely {{IPAc-en|ˈ|æ|z|ð|m|ə}})
* If the second consonant in a complex coda is voiced, so is the first
* An obstruent following {{IPA|/m/}} or {{IPA|/ŋ/}} in a coda must be [[homorganic]] with the nasal
* Two obstruents in the same coda must share [[Voice (phonetics)|voicing]] (compare ''ki'''ds''''' {{IPAc-en|k|ɪ|d|z}} with ''ki'''ts''''' {{IPAc-en|k|ɪ|t|s}})

===Ancient Greek===
{{main|Ancient Greek phonology#Phonotactics}}

Like English and some other [[Indo-European language]]s, Ancient Greek allowed onset clusters beginning in {{IPA|[s]}} which violate the SSP, such as {{lang|grc|στάδιον}} ([[stadion (unit)|stadion]]) or {{lang|grc|σφραγίς}} ([[sphragis (literary device)|sphragis]]: note that {{lang|grc|φ}} was originally pronounced {{IPA|[pʰ]}}). Ancient Greek also included initial consonant clusters such as {{IPA|[pt]}} in {{lang|grc|Πτολεμαῖος}} ([[Ptolemy]]) and {{IPA|[bd]}} in {{lang|grc|βδέλλιον}} ([[bdellion]]): unlike their borrowed versions in English, all the consonants were pronounced.<ref name="Goldstein">{{Harvnb|Goldstein|2014}}</ref>

===Japanese===
{{main|Japanese phonology#Phonotactics}}

The sound structure of [[Japanese language|Japanese]] is relatively straightforward compared to English.

==Notes and reference==

=== Notes ===
{{reflist}}

===References===
* Bailey, Todd M. & Hahn, Ulrike. 2001. Determinants of wordlikeness: Phonotactics or lexical neighborhoods? ''Journal of Memory and Language'' 44: 568–591.
* Coleman, John S. & Pierrehumbert, Janet. 1997. Stochastic phonological grammars and acceptability. ''Computational Phonology'' 3: 49–56.
* Frisch, S.; Large, N. R.; & Pisoni, D. B. 2000. Perception of wordlikeness: Effects of segment probability and length on processing non-words. ''Journal of Memory and Language'' 42: 481–496.
* Gathercole, Susan E. & Martin, Amanda J. 1996. Interactive processes in phonological memory. In ''Cognitive models of memory'', edited by Susan E. Gathercole. Hove, UK: Psychology Press.
* Hammond, Michael. 2004. Gradience, phonotactics, and the lexicon in English phonology. ''International Journal of English Studies'' 4: 1–24.
* Gaygen, Daniel E. 1997. Effects of probabilistic phonotactics on the segmentation of continuous speech. Doctoral dissertation, University at Buffalo, Buffalo, NY.
* Greenberg, Joseph H. & Jenkins, James J. 1964. Studies in the psychological correlates of the sound system of American English. ''Word'' 20: 157–177.
*{{cite encyclopedia |last=Goldstein |first=David |year=2014 |title=Phonotactics |encyclopedia=Encyclopedia of Ancient Greek Language and Linguistics |volume=3 |pages=96, 97 |publisher=Brill |via=[[academia.edu]] |url=https://www.academia.edu/4771043 |access-date=19 January 2015 }}
* {{cite book|last1=Laufer|first1=B.|year=1997 |chapter=What’s in a word that makes it hard or easy? Some intralexical factors that affect the learning of words |title=Vocabulary: Description, Acquisition and Pedagogy |isbn=9780521585514|publisher=Cambridge University Press|location=Cambridge|pages=140–155}}
* Luce, Paul A. & Pisoni, Daniel B. 1998. Recognizing spoken words: The neighborhood activation model. ''Ear and Hearing'' 19: 1–36.
* Newman, Rochelle S.; Sawusch, James R.; & Luce, Paul A. 1996. Lexical neighborhood effects in phonetic processing. ''Journal of Experimental Psychology: Human Perception and Performance'' 23: 873–889.
* Ohala, John J. & Ohala, M. 1986. Testing hypotheses regarding the psychological manifestation of morpheme structure constraints. In ''Experimental phonology'', edited by John J. Ohala & Jeri J. Jaeger, 239–252. Orlando, FL: Academic Press.
* Orzechowska, Paula; [[Richard Wiese (linguist)|Wiese, Richard]]. 2015; Preferences and variation in word-initial phonotactics: a multi-dimensional evaluation of German and Polish. ''Folia Linguistica'' 49: 439-486.
* Pitt, Mark A. & McQueen, James M. 1998. Is compensation for coarticulation mediated by the lexicon? ''Journal of Memory and Language'' 39: 347–370.
* Storkel, Holly L. 2001. Learning new words: Phonotactic probability in language development. ''Journal of Speech, Language, and Hearing Research'' 44: 1321–1337.
* Storkel, Holly L. 2003. Learning new words II: Phonotactic probability in verb learning. ''Journal of Speech, Language, and Hearing Research'' 46: 1312–1323.
* Vitevitch, Michael S. & Luce, Paul A. 1998. When words compete: Levels of processing in perception of spoken words. ''Psychological Science'' 9: 325–329.
* Vitevitch, Michael S. & Luce, Paul A. 1999. Probabilistic phonotactics and neighborhood activation in spoken word recognition. ''Journal of Memory and Language'' 40: 374–408.
* Vitevitch, Michael S.; Luce, Paul A.; Charles-Luce, Jan; & Kemmerer, David. 1997. Phonotactics and syllable stress: Implications for the processing of spoken nonsense words. ''Language and Speech'' 40: 47–62.
* Vitevitch, Michael S.; Luce, Paul A.; Pisoni, David B.; & Auer, Edward T. 1999. Phonotactics, neighborhood activation, and lexical access for spoken words. ''Brain and Language'' 68: 306–311.'


==External links==
* [http://www.iphod.com/ The Irvine Phonotactic Online Dictionary (IPhOD)]
* [http://phonotactics.anu.edu.au/ World Phonotactics Database] {{dead link|date=April 2025}}

[[Category:Phonotactics| ]]
[[Category:Phonology]]
[[Category:Mathematical linguistics]]