Debuccalization#Proto-Greek

{{IPA|/q/}} is debuccalized to {{IPA|/ʔ/}} in several Arabic varieties, such as northern Egyptian, Lebanese, western Syrian, and urban Palestinian dialects, partially also in Jordanian Arabic (especially by female speakers).{{Cite book|last=Bassiouney|first=Reem|title=Arabic Sociolinguistics|url=https://archive.org/details/arabicsociolingu00bass|url-access=limited|publisher=Georgetown University Press|year=2009|isbn=978-1-58901-573-9|location=Washington, DC|pages=[https://archive.org/details/arabicsociolingu00bass/page/n175 158]-161}} The Maltese language, which was originally an Arabic dialect, also shows this feature.

Most English-speakers in England and many speakers of American English debuccalize {{IPA|/t/}} to a glottal stop {{IPA|[ʔ]}} in two environments: in word-final position before another consonant (American English IPA)

• get ready {{IPA|[ˈɡɛʔˈɹɛɾi]}}
• not much {{IPA|[ˈnɑʔˈmʌtʃ]}}
• not good {{IPA|[ˈnɑʔˈɡʊd̚]}}
• it says {{IPA|[ɪʔˈsɛz]}}

Before a syllabic {{IPA|[n̩]}} following {{IPA|/l/}}, {{IPA|/r/}}, {{IPA|/n/}}, a vowel, or a diphthong. The {{IPA|/t/}} may then also be nasally released. (American English IPA)

• Milton {{IPA|[ˈmɪlʔn̩]}}
• Martin {{IPA|[ˈmɑɹʔn̩]}}
• mountain {{IPA|[ˈmæʊnʔn̩]}}
• cotton {{IPA|[ˈkʰɑʔn̩]}}
• Latin {{IPA|[ˈlæʔn̩]}}
• Layton {{IPA|[ˈleɪʔn̩]}}

In Cockney English, {{IPA|/t/}} is often realized as a glottal stop {{IPA|[ʔ]}} between vowels, liquids, and nasals (notably in the word bottle), a process called t-glottalization.

The German ending -en is commonly realized as an assimilated syllabic nasal. Preceding voiceless stops are then glottally released: {{lang|de|Latten}} {{IPA|[ˈlat͡ʔn̩]}} ('laths'), {{lang|de|Nacken}} {{IPA|[ˈnak͡ʔŋ̍]}} ('nape of the neck'). When such a stop is additionally preceded by a homorganic sonorant, it tends to be debuccalized entirely and create the clusters {{IPA|[mʔm̩, lʔn̩, nʔn̩, ŋʔŋ̍]}}. For example, {{lang|de|Lumpen}} {{IPA|[ˈlʊmʔm̩]}} ('rag'), {{lang|de|Banken}} {{IPA|[ˈbaŋʔŋ̍]}} ('banks').

Voiced stops are not usually debuccalized. However, many Upper German and East Central German dialects merge voiced and unvoiced stops at least word-internally, and the merged consonants may be debuccalized. For example, in Bavarian, both {{lang|bar|Anten}} ('ducks') and {{lang|bar|Anden}} ('Andes') are pronounced {{IPA|[ˈɑnʔn̩]}}. Speakers are often unaware of that.

However, Standard German spoken in Luxembourg often lacks syllabic sonorants under the influence of Luxembourgish, so that {{lang|de|-en}} is pronounced {{IPA|[ən]}}, rather than {{IPA|[n̩]}} or {{IPA|[ŋ̍]}}.{{cite book|author1=Dudenredaktion|last2=Kleiner|first2=Stefan|last3=Knöbl|first3=Ralf|year=2015|orig-year=First published 1962|title=Das Aussprachewörterbuch|url=https://books.google.com/books?id=T6vWCgAAQBAJ|language=de|edition=7th|location=Berlin|publisher=Dudenverlag|isbn=978-3-411-04067-4|page=39}}{{cite journal|last1=Gilles|first1=Peter|last2=Trouvain|first2=Jürgen|year=2013|title=Luxembourgish|journal=Journal of the International Phonetic Association|volume=43|issue=1|pages=70–71|doi=10.1017/S0025100312000278|url=https://www.cambridge.org/core/services/aop-cambridge-core/content/view/3C9FB295A261FD6F28D694252B06B4A3/S0025100312000278a.pdf/luxembourgish.pdf|doi-access=free}}

In both languages, syllable-final -k is either realized as {{IPA|[k]}} or {{IPA|[ʔ]}}.

Debuccalization is very common in parts of Sulawesi. Especially in the South Sulawesi branch, most languages have turned word-final *t and *k into a glottal stop.{{Cite thesis |last=Mills |first=Roger Frederick |title=Proto South Sulawesi and Proto Austronesian Phonology |date=1975a |type=Ph.D. dissertation |publisher=University of Michigan |hdl=2027.42/157431 |language=en}}

In every Gorontalic language except Buol and Kaidipang, *k was replaced by a glottal stop, and lost altogether in word-initial position: *kayu → Gorontalo ayu {{gloss|wood}}, *konuku → olu'u {{gloss|fingernail}}. However, if it followed *ŋ, then *k voiced into g in Gorontalo (*koŋkomo → onggomo {{gloss|handful}}).{{cite journal |last1=Sneddon |first1=James N. |last2=Usup |first2=Hunggu Tadjuddin |year=1986 |title=Shared sound changes in the Gorontalic language group: Implications for subgrouping |journal=Bijdragen tot de Taal-, Land- en Volkenkunde |volume=142 |issue=4 |pages=407–26 |jstor=27863783 |doi=10.1163/22134379-90003347 |doi-access=free }}

Debuccalization is also common in the Sangiric branch. In Sangir and Bantik, all final voiceless stops were reduced into ʔ (*manuk → manu' "bird"). Also in Ratahan, final *t became ʔ (*takut → taku' "to fear"). In Talaud, all instances of Proto-Sangiric *k were debuccalized into ʔ except when following *ŋ (*kiki → i'i "to bite", but *beŋkol → bengkola "bent"). Other newer instances of k resulted from *R when geminated or being word-final (ʐ elsewhere), e.g. *bəRu → bakku "new", *bibiR → biwikka "lip", *bəŋaR → bangngaka "molar".{{Cite journal |last=Sneddon |first=James N. |author-link=James Sneddon |year=1984 |title=Proto-Sangiric and the Sangiric languages |location=Canberra |journal=Pacific Linguistics |doi=10.15144/PL-B91 |doi-access=free}}

Many Polynesian languages lost the original glottal stop *ʔ of their ancestor Proto-Polynesian, but then debuccalized other consonants into a glottal stop {{IPA|/ʔ/}}. This applied to different consonants depending on the language, for example:See p.93-95 of: {{cite book |last1=Charpentier |first1=Jean-Michel |first2=Alexandre |last2=François |author-link2=Alexandre François (linguist) |year=2015 |title=Atlas Linguistique de Polynésie Française — Linguistic Atlas of French Polynesia |language=fr, en |publisher=Mouton de Gruyter & Université de la Polynésie Française |isbn=978-3-11-026035-9 |ref=atlas |url=http://alex.francois.online.fr/AF-Atlas-blurbs_e.htm }}

• Samoan {{IPA|/ʔ/}} < {{abbrlink|PPn|Proto-Polynesian language}} *k
• Tahitian {{IPA|/ʔ/}} < PPn *k, *ŋ
• Marquesan {{IPA|/ʔ/}} < PPn *l, *r
• Mangarevan {{IPA|/ʔ/}} < PPn *f, *s
• Rurutu {{IPA|/ʔ/}} < PPn *k, *ŋ, *f, *s
• Hawaiian {{IPA|/ʔ/}} < PPn *k, *l, *r.

Older {{IPA|/ɡ/}} was spirantized and later debuccalized in languages such as Belarusian, the Czech–Slovak languages, Ukrainian, and Upper Sorbian, e.g. Serbian bog, Russian box, Czech bůh, Ukrainian bih.

In some varieties of Scots and Scottish English, particularly on the West Coast, a non word-final {{IPA|/θ/}} th shifted to {{IPA|[h]}}, a process called th-debuccalization. For example, {{IPA|/θɪn/}} is realized as {{IPA|[hɪn]}}.

Pre-pausally, {{IPA|/t/}} may be debuccalized to [h], eg. it, lot, that, what pronounced {{IPA|[ɪh, lɒh, d̪ah, wɒh]}}.

In Proto-Greek, {{IPA|/s/}} shifted to {{IPA|[h]}} initially and between sonorants (vowels, liquids, and nasals).

• Proto-Indo-European {{wikt-lang|ine-x-proto|*septḿ̥}} → Proto-Greek {{lang|grk-x-proto|*heptə́}} → Ancient Greek {{Transliteration|grc|heptá}} ({{wikt-lang|grc|ἑπτά}}) "seven" (vs. Latin {{wikt-lang|la|septem}})

Intervocalic {{IPA|/h/}} had been lost by the time of Ancient Greek, and vowels in hiatus were contracted in the Attic dialect.

• post-PIE *ǵénesos → Proto-Greek {{lang|grk-x-proto|*génehos}} → Ionic {{Transliteration|grc|géneos}} ({{lang|grc|γένεος}}) : Attic {{Transliteration|grc|génous}} ({{lang|grc|γένους}}) "of a race"

Before a liquid or nasal, an {{IPA|/h/}} was assimilated to the preceding vowel in Attic-Ionic and Doric and to the following nasal in Aeolic. The process is also described as the loss of {{IPA|/h/}} and the subsequent lengthening of a vowel or consonant, which kept the syllable the same length (compensatory lengthening).

• PIE {{wikt-lang|ine-x-proto|*h₁ésmi}} → Proto-Greek {{lang|grk-x-proto|*ehmi}} → Attic-Ionic {{Transliteration|grc|ēmí}} ({{lang|grc|εἰμί}}) : Aeolic {{Transliteration|grc|émmi}} ({{lang|grc|ἔμμι}}) "I am"

In Sanskrit, {{IPA|/s/}} becomes {{IPA|[h]}} (written {{Transliteration|sa|ḥ}} in transliteration) before a pause: e.g. {{Transliteration|sa|kā́mas}} ('desire') becomes {{Transliteration|sa|kā́maḥ}}.

Additionally, the Proto-Indo-European aspirated voiced palato-velar *ǵʰ {{IPA|[ɟʱ]}} became {{IPA|[ɦ]}} through successive affrication, assibilation and debuccalization: e.g. {{Transliteration|ine|*bʰeh₂ǵʰús}} "arm" becomes Sanskrit {{Transliteration|sa|bāhúḥ}}.

In many Eastern Bengali dialects, the voiceless palato-alveolar sibilant {{IPA|/ʃ/}} can become debuccalized to glottal {{IPA|[h]}} or {{IPA|[ɦ]}}, e.g. {{IPA|/ʃälä/}} "wife's brother" is {{IPA|[ɦälä]}}, and {{IPA|/ʃägoɾ/}} "sea" is {{IPA|[ɦä(g)oɾ]}}. The tenuis and aspirated forms of the labial stop {{IPA|/p/, /pʰ/}} and velar stop {{IPA|/k/, /kʰ/}} can get lenited to {{IPA|/ɸ/}} and {{IPA|/x/}} respectively, but also be further debuccalized to {{IPA|[h]}} or {{IPA|[ɦ]}}, e.g. {{IPA|/pägol/}} "mad" is {{IPA|[ɦägol]}} and {{IPA|/pʰokiɾ/ ~ /ɸokiɾ/}} "beggar, faqir" is {{IPA|[ɸoɦiɾ]}}. In some cases, even the glottal fricative is dropped, e.g. {{IPA|/äʃilo/}} "(he / she / it) came" is {{IPA|[äi̯lo]}}.

{{main|Spanish dialects and varieties#Debuccalization of coda /s/|l1=Spanish dialects and varieties: Debuccalization of coda /s/}}

A number of Spanish dialects debuccalize {{IPA|/s/}} to {{IPA|[h]}} or {{IPAblink|ɦ}} at the end of a syllable or intervocalically in certain instances.

In many varieties of Galician, as well as in Galician-influenced Spanish, the phoneme {{IPA|/ɡ/}} may debuccalize ({{lang|gl|gheada}}) to {{IPAblink|ħ}} in most or all instances; {{IPAblink|x}} and {{IPAblink|h}} are also possible realizations. There is also an inverse hypercorrection process of older or less educated Galician speakers replacing the phoneme {{IPAslink|x}} of the Spanish language with {{IPA|[ɡ]}}, which is called {{lang|gl|gueada}}.

Portuguese is much less affected by debuccalization, but it is especially notable in its Brazilian variety.

Throughout Brazil, the phoneme {{IPA|/ʁ/}} (historically an alveolar trill {{IPA|/r/}} that moved to an uvular position) has a rather long inventory of allophones: {{IPA|[{{IPAplink|r}} {{IPAplink|ʂ|ɻ̝̊}} {{IPAplink|ç}} {{IPAplink|x}} {{IPAplink|ɣ}} {{IPAplink|χ}} {{IPAplink|ʁ}} {{IPAplink|ʀ}} {{IPAplink|ħ}} {{IPAplink|h}} {{IPAplink|ɦ}}]}}. Only {{IPAblink|ɣ}} is uncommon. Few dialects, such as {{lang|pt|sulista}} and {{lang|pt|fluminense}}, give preference to voiced allophones; elsewhere, they are common only as coda, before voiced consonants.

In such dialects, especially among people speaking an educated variety of Portuguese, it is usual for the rhotic coda in the syllable rhyme to be an alveolar tap, as in European Portuguese and many registers of Spanish, or to be realized as {{IPAblink|χ}} or {{IPAblink|x}}. In the rest of the country, it is generally realized as {{IPAblink|h}}, even by speakers who either do not normally use that allophone or delete it entirely, as is common in the vernacular.

However, in some {{lang|pt|mineiro}}- and {{lang|pt|mineiro}}-influenced {{lang|pt|fluminense}} rural registers, {{IPA|[h]}} is used but as an allophone of {{IPAslink|l}} (rhotic consonants are most often deleted), a {{lang|pt|mar-mal}} merger, instead of the much more common and less-stigmatized L-vocalization#Portuguese characteristic of all Brazilian urban centers except for those bordering Mercosur countries, where coda {{IPAblink|ɫ}} was preserved, and the entire North and Northeast regions. Its origin is the replacement of indigenous languages and {{lang|pt|línguas gerais}} by Portuguese,{{Citation needed|date=May 2016}} which created {{IPAblink|ɹ}}, {{IPAblink|ɻ}} and r-colored vowel as allophones of both {{IPA|/ɾ/}} (now mostly {{IPA|/ʁ/}}) and {{IPA|/l/}} (now mostly {{IPA|[{{IPAplink|w|u̯}} ~ {{IPAplink|ʊ|ʊ̯}}]}}) phonemes in the coda since Native Brazilians could not easily pronounce them (caipira dialect).{{Citation needed|date=May 2016}} The later Portuguese influence from other regions made those allophones become rarer in some areas, but the {{lang|pt|mar-mal}} merger remained in a few isolated villages and towns.

Finally, many {{lang|pt|fluminense}} registers, especially those of the poor and of the youth, most northern and northeastern dialects, and, to a much minor degree, all other Brazilian dialects, debuccalize {{IPA|/s/}} but less so than in Spanish. However, a {{lang|pt|mar-mas}} merger or even a {{lang|pt|mar-mais}} merger occurs: {{lang|pt|mas mesmo assim}} "but even so" or {{lang|pt|mas mesma, sim}} "though, right, the same (f) one" {{IPA|[mɐɦ ˈmeɦmə ˈsĩ]}}; {{lang|pt|mais}} light "lighter, more slim", or also "less caloric/fatty" {{IPA|[ˈmaɦ ˈlajtɕ]}}; {{lang|pt|mas de mim, não}} "but from me, no" or {{lang|pt|mais de mim, não}} "not more from me" {{IPA|[ˈmaɦ dʑi ˈmĩ ˈnɜ̃w]}}. A coda rhotic in the Brazilian dialects in the Centro-Sul area is hardly ever glottal, and the debuccalized {{IPA|/s/}} is unlikely to be confused with it.

In the Moldavian dialect of Romanian, {{IPAslink|f}} is debuccalized to {{IPA|[h]}} and so, for example, {{lang|ro|să fie}} becomes {{lang|ro|să hie}}. The same occurred in Old Spanish, Old Gascon, and still occurs in Sylheti.

{{main|Scottish Gaelic phonology#Lenition and spelling|Irish initial mutations|l1=Scottish Gaelic phonology: Lenition and spelling}}

In Scottish and Irish Gaelic, s and t changed by lenition to {{IPA|[h]}}, spelled sh and th.

Inscription in Faliscan from the 4th century BC on show occasional debuccalization of {{IPA|/f/}} to {{IPA|/h/}} (e.g. hileo : Latin filius). Whether the shift is displayed in the inscriptions is highly irregular, with some forms even showing an ostensibly opposite shift of written f in place of an expected h (e.g. fe : Latin hic), possibly by means of hypercorrection.{{cite journal|title=On the Problematic f/h Variation in Faliscan|author=Rex E. Wallace|author2=Brian D. Joseph|journal=Glotta|page=90|jstor=40266879|year=1991}}

In several Malay dialects in the peninsular, final -s is realized as {{IPA|[h]}}.

In the Batak branch, all southern languages (but not the northern ones including Karo), have debuccalized *k into h, except when word-final or followed *ŋ. Both Angkola and Mandailing have restored k within the sequence hVhV (Angkola kehe, Mandailing ke, but Toba hehe), or when following a consonant in Mandailing (ala "scorpion" → parkalahan). Mandailing, however has also further deleted *h (*kalak → alak "person"), except in the sequence -aha- (dahan "mushroom", not *dan).Adelaar, K. A. (1981). "Reconstruction of Proto-Batak Phonology". In Robert A. Blust (ed.), Historical Linguistics in Indonesia: Part I, 1–20. Jakarta: Universitas Katolik Indonesia Atma Jaya.

Polynesian languages commonly reflect debuccalization not only into a glottal stop {{IPA|/ʔ/}}, but also into a glottal fricative {{IPA|/h/}}. The exact distribution depends on the language:

• most languages reflect a regular change {{abbrlink|PPn|Proto-Polynesian language}} *s > {{IPA|/h/}}
• in several languages, the outcome of PPn *f is irregular across the lexicon, with no obvious conditioning:
  PPn *f > Tahitian {{IPA|/f/}}, {{IPA|/h/}}; Māori {{IPA|/ɸ/}}, {{IPA|/h/}}; Hawaiian {{IPA|/w/}}, {{IPA|/h/}}, etc.

{{main|Yoruba language#Yoruba phonology|Yoruboid languages|Proto-Yoruboid language}}

Debuccalization occurs extensively within the dialectal continuum of Yoruboid languages, particularly among the Olukumi language, Igala language, the Northeast Yoruba dialect known as Owe, and Southeastern dialects of the Yoruba language, such as Ikale. Many of these shifts came from Proto-Yoruboid language (or its descendant language, Proto-Edekiri), and descendant languages shifted from {{IPA|/s/}} to {{IPA|/h/}}. In other cases shifts from {{IPA|/f/}} to {{IPA|/h/}} also occur from Proto-Yoruboid to Standard Yoruba. Many other alternatives shift from {{IPA|/s/}} to {{IPA|/r/}}, but it is unclear if that process is associated with the debuccalization occurring.

• Proto-Yoruboid {{wikt-lang|yo|*sì}} → Igala {{wikt-lang|igl|hì}}, Proto-Edekiri *sè → Owe {{wikt-lang|yo|hè}}, Olukumi hè, Ikale {{wikt-lang|yo|hè}} "to cook" (vs. Standard Yoruba {{wikt-lang|yo|sè}})
• Proto-Yoruboid {{wikt-lang|yo|*è-so}} → Igala {{wikt-lang|igl|èho}}, Proto-Edekiri *è-ho → Owe {{wikt-lang|yo|èho}}, Olukumi èhojin "fruit, seed (something that is sowed)" (vs. Standard Yoruba {{wikt-lang|yo|èso}})
• Proto-Edekiri *V̀-sʊ̃ → Ikale {{wikt-lang|yo|ùhọngbẹ́}}, Olukumi ùhọn "ground squirrel" (vs. Ekiti Yoruba {{wikt-lang|yo|ụ̀sụn}})
• Proto-Yoruboid {{wikt-lang|yo|*á-folo}} → Igala {{wikt-lang|igl|áfolo}} (vs. Standard Yoruba {{wikt-lang|yo|ehoro}})

Debuccalization also occurs in other Volta-Niger languages, including Igbo, the Ayere-Ahan languages, and the Edo.

In Early Modern Japanese, the labial fricative {{IPA|[ɸ]}} (derived from Old Japanese *{{IPA|/p/}}) was debuccalized to {{IPA|[h]}} when followed by one of the vowels {{IPA|/a, o, e/}}. (It remained labial {{IPA|[ɸ]}} before {{IPA|/u/}}, and was palatalized to {{IPA|[ç]}} before {{IPA|/i, j/}}.)

In old Kannada at around 10th-14th century, most of the initial {{IPA|/p/}} debuccalized into a {{IPA|/h/}} e.g. OlKn. pattu, MdKn. hattu "ten".{{sfnp|Krishnamurti|2003|p=120}}

All coda consonants in Slavey must be glottal. When a non-glottal consonant would otherwise be positioned in a syllable coda, it debuccalizes to {{IPA|[h]}}:{{Harvcoltxt|Rice|1989|p=144,150}}

• {{IPA|/ts’ad/}} → {{IPA|[ts’ah]}} ('hat')
• {{IPA|/xaz/}} → {{IPA|[xah]}} ('scar')
• {{IPA|/tl’uɮ/}} → {{IPA|[tl’uh]}} ('rope')

Debuccalization can be a feature of loanword phonology. For example, debuccalization can be seen in Indonesian loanwords into Selayar.{{Harvcoltxt|O'Brien|2012|p=28}}

{{reflist}}

• {{Cite book |title=The Dravidian Languages |last=Krishnamurti |first=Bhadriraju |year=2003 |publisher=Cambridge University Press |isbn=978-0-521-77111-5 |url=https://books.google.com/books?id=54fV7Lwu3fMC}}
• {{citation

|last=O'Brien

|first=Jeremy Paul

|year=2012

|title=An experimental approach to debuccalization and supplementary gestures

|url=http://escholarship.org/uc/item/1cm694ff

}}

• {{citation

|last=Rice

|first=Keren

|year=1989

|title=A Grammar of Slave

|url=https://www.degruyter.com/view/title/10997

}}

• [https://books.google.com/books?id=wx-4wmfQbVMC&dq=debuccalization+1990&pg=PA123 "Debuccalization"] (Chapter 4 of Paul D. Fallon's The Synchronic and Diachronic Phonology of Ejectives [Routledge, 2001]) gives many other terms that have been proposed for the phenomenon.
• [http://people.ucsc.edu/~jpobrien/papers/obrien_qe.pdf Debuccalization and supplementary gestures]{{Dead link|date=July 2019 |bot=InternetArchiveBot |fix-attempted=yes }}

Category:Phonology