Product Code Database
The phonology of the Korean language covers the language's distinct, meaningful sounds (19 and 7 in the standard Seoul dialect) and the rules governing how those sounds interact with each other. This article is a technical description of the phonetics and phonology of Korean language. Unless otherwise noted, statements in this article refer to South Korean standard language based on the Gyeonggi dialect.
For each plosive and affricate, there is a three-way contrast between unvoiced segments, which are distinguished as plain, tense, and aspirated.
Korean syllable structure is maximally CGVC, where G is a glide . (There is a unique off-glide diphthong in the character 의 that combines the sounds and creating ). Any consonant except may occur initially, but only may occur finally. Sequences of two consonants may occur between vowels.
| + Consonant phonemes ! colspan="2" | ! Bilabial ! Alveolar ! Alveolo-palatal/Palatal ! Velar consonant ! Glottal |
| + Example words for consonant phonemes ! rowspan="2" | Example | ||
| 불]] | 'fire' or 'light' | ||
| 뿔]] | 'horn' | ||
| 풀]] | 'grass' or 'glue' | ||
| 물]] | 'water' or 'liquid' | ||
| 달]] | 'moon' or 'month' | ||
| 딸]] | 'daughter' | ||
| 탈]] | 'mask' or 'trouble' | ||
| 날]] | 'day' or 'blade' | ||
| 자다]] | 'to sleep' | ||
| 짜다]] | 'to squeeze' or 'to be salty' | ||
| 차다]] | 'to kick' or 'to be cold' | ||
| 기]] | 'energy' | ||
| 끼]] | 'talent' or 'meal' | ||
| 키]] | 'height' | ||
| 방]] | 'room' | ||
| 살]] | 'flesh' | ||
| 쌀]] | 'uncooked grains of rice' | ||
| 바람]] | 'wind' or 'wish' | ||
| 발]] | 'foot' | ||
| 하다]] | 'to do' | ||
are voiced between sonorants (including all vowels and certain consonants) but voiceless elsewhere. Among younger generations, they may be just as aspirated as in initial position; the primary difference is that vowels following the plain consonants [[carry low tone|tonogenesis]].
are strongly aspirated, more so than English voiceless stops. They generally do not undergo intervocalic voicing, but a 2020 study reports that it still occurs in around 10 to 15% of cases. It is more prevalent among older male speakers who have aspirated stops voiced in as much as 28% of cases.
They are produced with a partially constricted glottis and additional subglottal pressure in addition to tense vocal tract walls, laryngeal lowering, or other expansion of the larynx.
An alternative analysis proposes that the "tensed" series of sounds are (fundamentally) regular voiceless, unaspirated consonants: the "lax" sounds are voiced consonants that become devoiced initially, and the primary distinguishing feature between word-initial "lax" and "tensed" consonants is that initial Lax vowel sounds cause the following vowel to assume a low-to-high pitch contour, a feature reportedly associated with voiced consonants in many Asian languages (such as Shanghainese), whereas tensed (and also aspirated) consonants are associated with a uniformly high pitch.
Vowels before tense consonants (as well as aspirated) tend to be shorter than before lax stops.
The Gyeongsang dialect is known for realization of tense ㅆ as plain ㅅ .
The analysis of as phonologically plain or aspirated has been a source of controversy in the literature. Similarly to plain stops, it shows moderate aspiration word-initially but no aspiration word-medially. It also often undergoes intervocalic voicing. But similar to aspirated stops, it triggers high pitch in the following vowel.
Word-initial aspiration, intervocalic voicing, and higher pitch of the following vowels are shared qualities in Korean fricatives and .
ㅁ, ㄴ tend to be denasalization word-initially.
ㅇ does not occur in initial position, reflected in the way the Hangul jamo ㅇ has a different pronunciation in the initial position to the final position. These were distinguished when Hangul was created, with the jamo ㆁ with the upper dot and the jamo ㅇ without the upper dot; these were then conflated and merged in both the North Korean and South Korean standards. can technically occur syllable-initially, as in 명이, which is written as , but pronounced as .
ㄹ is an alveolar flap between vowels or between a vowel and an . It is or at the end of a word, before a consonant other than , or next to another ; in these contexts, it is palatalized to before and before palatal consonant allophones. There is free variation at the beginning of a word, where this phoneme tends to become before most vowels and silent before , but it is commonly in English loanwords. Geminate is realized as , or as before .
In native Korean words, ㄹ does not occur word initially, unlike in Chinese loans (Sino-Korean vocabulary). In South Korea, it is silent in initial position before and , pronounced before other vowels, and pronounced only in compound words after a vowel. The prohibition on word-initial is called the "initial sound law" or (두음법칙). Initial is spelled with ㄹ in North Korea, but is often pronounced the same way as it is in South Korea.
This rule also extends to ㄴ in many native and all Sino-Korean words, which is also lost before initial and in South Korean; again, North Korean preserves the phoneme there.
In both countries, initial in words of foreign origin other than Chinese is pronounced . Very old speakers may pronounce word-initial as even in Western loanwords, e.g. in "lighter" 라이터 .
When pronounced as an alveolar flap , ㄹ is sometimes derived from through a historical lenition process, such as in the verb 듣다 which turns into 들어요.
The features of consonants are summarized in the following table.
| + Features of consonants (2025). 9780824826017, University of Hawaii Press. ISBN 9780824826017
! Consonant class
! Voice
! Tension
! Aspiration
! Pitch of following vowel |
When such a sequence is followed by a consonant, the same reduction takes place, but a trace of the lost consonant may remain in its effect on the following consonant. The effects are the same as in a sequence between vowels: an elided obstruent will leave the third consonant fortis, if it is a stop, and an elided will leave it aspirated. Most conceivable combinations do not actually occur; a few examples are = , = , = , = , = , = ; also = , as has no effect on a following , and = , with the dropping out.
When the second and third consonants are homorganic obstruents, they merge, becoming fortis or aspirate, and, depending on the word and a preceding , might not elide: is .
An elided has no effect: = , = , = , = , = , = , = , = , = , = , = .
The table below is out of alphabetical order to make the relationships between the consonants explicit:
All (stops, affricates, fricatives) become stops with no audible release at the end of a word: all coronals collapse to , all labials to , and all velars to . Final ㄹ is a lateral or .
ㅈ, ㅊ, ㅉ are pronounced in Seoul, but typically pronounced in Pyongyang. Similarly, are palatalized as before in Seoul. In Pyongyang they remain unchanged. This pronunciation may be also found in Seoul Korean among some speakers, especially before back vowels.
As noted above, initial is silent in this palatalizing environment, at least in South Korea. Similarly, an underlying or at the end of a morpheme becomes a phonemically palatalized affricate or , respectively, when followed by a word or suffix beginning with or (it becomes indistinguishable from an underlying ), but that does not happen within native Korean words such as "where?".
is more affected by vowels, often becoming an affricate when followed by or : , . The most variable consonant is , which becomes a palatal before or , a [[velar|velar consonant]] before , and a bilabial before , and .(2025). 9781107672680, Cambridge University Press. ISBN 9781107672680
| + Allophones of consonants before vowels ! ! width="75px" | ! width="75px" | ! width="75px" | ! width="75px" |
In many morphological processes, a vowel before another vowel may become the semivowel . Likewise, and , before another vowel, may reduce to . In some dialects and speech registers, the semivowel assimilates into a following or and produces the front rounded vowels and .
Korean also features regressive (anticipatory) assimilation, where a consonant tends to assimilates in manner but not in place of articulation. For example, Obstruents become nasal stops before nasal stops (which, as just noted, includes underlying ), but do not change their position in the mouth. Velar stops (that is, all consonants pronounced in final position) become ; coronals () become , and labials () become . For example, (한국말) is pronounced (한궁말) (phonetically ).
Before the fricatives , coronal obstruents assimilate to a fricative, resulting in a geminate. That is, is pronounced (). A final assimilates in both place and manner, so that is pronounced as a geminate (and, as noted above, aspirated if C is a stop). The two coronal sonorants, and , in whichever order, assimilate to , so that both and are pronounced .
There are lexical exceptions to these generalizations. For example, voiced consonants occasionally cause a following consonant to become fortis rather than voiced; this is especially common with and as and , but is also occasionally seen with other sequences, such as (), () and ().
Incorrect application of these phonological rules, such as improper nasalization or assimilation, can significantly impair intelligibility for native speakers. This highlights the need for comprehensive instruction in these rules to ensure clear and effective communication. For instance, failing to apply nasalization rules correctly or not recognizing assimilation patterns can lead to pronunciations that are difficult for native speakers to understand, underscoring the importance of precise phonological training.
| + Phonetic realization (before ) of underlying consonant sequences in Korean |
The resulting geminate obstruents, such as , , , and (that is, , , , and ), tend to reduce (, , , ) in rapid conversation. Heterorganic obstruent sequences such as and may, less frequently, assimilate to geminates (, ) and also reduce to (, ).
These sequences assimilate with following vowels the way single consonants do, so that for example and palatalize to (that is, ) before and ; and affricate to and before ; , , and palatalize to and across morpheme boundaries, and so on.
Hangul orthography does not generally reflect these assimilatory processes, but rather maintains the underlying morphology in most cases.
| + Seoul Korean monophthongs ! rowspan="2" | ! Front vowel !Central vowel ! colspan="3"Back vowel |
| + Pyongan dialect monophthongs ! rowspan="2" | ! Front vowel !Central vowel ! colspan="2"Back vowel |
The distinction between and is lost in South Korean dialects—both are most commonly realized as , but some older speakers still retain the difference; as for North Korean, some works report the distinction to be robust. However, the data from one study suggests that while younger KCTV anchors try to produce them more or less distinctly, it is not clear whether that is learned or natural pronunciation, as they do so inconsistently. Notably, older anchor Ri Chun-hee and even Kim Jong-un both have and merged.
In Gyeonggi dialect, is produced higher than , while in North Korean dialects the two are comparable in height, and is more fronted. In Gyeongsang dialect, and once have merged into in speech of older speakers, but they are distinct among young and middle-aged Daegu residents (they actually have the same vowels as Seoulites due to influence from Standard Korean).
In Seoul, is fronted, while is raised, and both are almost the same height, though is still more rounded. Due to this, alternative transcriptions like or for , and or for are proposed. In both varieties, is fronted away from , and in North Korean it is also lower, shifting more towards .
Korean used to have two additional phonemes, ㅚ and ㅟ, but they are replaced by the diphthongs and by the majority of speakers.
Middle Korean had an additional vowel phoneme denoted by ᆞ, known as (literally "lower a"). The vowel merged with in all mainland varieties of Korean, but remains distinct in Jeju language where it is pronounced .
| + Diphthongs, disregarding length ! IPA !! Hangul !! colspan="3" | Example |
| ㅖ | 'budget' |
| ㅒ | 'story' |
| ㅑ | 'baseball' |
| ㅛ | 'teacher' |
| ㅠ | 'glass' |
| ㅕ | 'here' |
| ㅟ | 'back' |
| ㅞ | 'chest' or 'box' |
| ㅙ | 'why' |
| ㅘ | 'fruit' |
| ㅝ | 'what' |
| ㅢ | 'doctor' |
In modern pronunciation, merges into after a consonant. Some analyses treat as a central vowel and thus the marginal sequence as having a central-vowel onset, which would be more accurately transcribed or .
Modern Korean has no falling diphthongs, with sequences like being considered as two separate vowels in hiatus. Middle Korean had a full set of diphthongs ending in , but these monophthongized into modern-day front vowels in Early Modern Korean (, , , , ). This is the reason why the hangul letters ㅐ, ㅔ, ㅚ etc. are represented as back vowels plus .
The sequences do not occur, and it is not possible to write them using standard hangul. The semivowel occurs only in the diphthong , and is prone to being deleted after a consonant.
Vowel length is a remnant of rising tone, first emerging in Middle Korean. It was preserved only in initial syllables and was often neutralized, particularly in the following cases:
It has disappeared gradually among younger speakers, but some middle-aged speakers are still aware of it and can still produce it in conscious speech. The long–short merger has had two main aspects. The first is phonetic: The duration of long vowels in relation to short ones has reduced by a lot (from 2.5:1 in the 1960s to 1.5:1 in the 2000s). Some studies suggest that the length of all vowels is dependent on one's age (older speakers seem to exhibit a slower speech rate, and even their short vowels are produced relatively longer than those of younger speakers). The second aspect is lexical: The subset of words produced with long vowels has gotten smaller. Long vowels tend to be reduced most frequently in high-frequency words.
| + Vowel phonemes with length distinction ! IPA !! Hangul !! colspan="3" | Example |
| ㅣ | 'hunger' |
| 'market' | |
| ㅔ | 'pillow' |
| 'to cut' | |
| ㅐ | 'pear' |
| 'double' | |
| ㅏ | 'horse' |
| 'word, language' | |
| ㅗ | 'barley' |
| 'salary' | |
| ㅜ | 'eye' |
| 'snow' | |
| ㅓ | 'punishment' |
| 'bee' | |
| ㅡ | 'seniors' |
| 'food' | |
| ㅚ | 'church' |
| 'overcoat' | |
| ㅟ | 'mouse' |
| 'ghost' |
| + Korean vowel harmony | ||||||
| (ㆎ ) | ||||||
| ㅢ | ||||||
| Obsolete and dialectal sounds in parentheses. | ||||||
There are three classes of vowels in Korean: "positive", "negative", and "neutral". The vowel ㅡ (eu) is considered both partially neutral and partially negative. The vowel classes loosely follow the negative and positive vowels; they also follow orthography. Exchanging positive vowels with negative vowels usually creates different nuances of meaning, with positive vowels representing diminutives and negative vowels representing exaggeration:
As early as 2004, researchers have posited that pitch and voicing are the actual distinguishing features between modern Korean consonants rather than the standard but seemingly ill-defined "tense" vs "lax" theory. A 2013 study by Kang Yoon-jung and Han Sung-woo which compared voice recordings of Seoul speech from 1935 and 2005 found that in recent years, lenis consonants (ㅂㅈㄷㄱ), aspirated consonants (ㅍㅊㅌㅋ) and fortis consonants (ㅃㅉㄸㄲ) were shifting from a distinction via voice onset time to that of pitch change, and suggests that the modern Seoul dialect is currently undergoing tonogenesis. Kim Mi-Ryoung (2013) notes that these sound shifts still show variations among different speakers, suggesting that the transition is still ongoing. Cho Sung-hye (2017) examined 141 Seoul dialect speakers, and concluded that these pitch changes were originally initiated by females born in the 1950s, and has almost reached completion in the speech of those born in the 1990s. On the other hand, Choi Ji-youn et al. (2020) disagree with the suggestion that the consonant distinction shifting away from voice onset time is due to the introduction of tonal features, and instead proposes that it is a prosodically conditioned change.
Some changes are still ongoing. They depend on age and gender, the speech of young females tends to be most innovative, while old males are phonologically conservative.
|
|
