What Pronunciations the Consonants are NOT
The Tips and Notes for Alphabet 2 does a good job introducing what the consonants sounds are in some technical descriptions (using the International Phonetic Alphabet) and some layman descriptions (like describing aspirates as “adding a puff of air”). Of course, going by the comments I see in my inbox, it seems that people need a little more elaboration or another angle to understand the nature of some consonants. Many of these comments lean on romanization instead. Romanization is helpful as long as one understands the limitations involved and the potential for misdirection. But the Anglo-centric baggage that frequently comes with romanization also create a persistent distortion in understanding what the Korean consonants really are—especially when well-meaning comments use romanization to explain pronunciation.
Some stuff I’ve seen elsewhere on the web:
I’m going to narrow the scope of my case by first excluding some consonants that I do not think are relevant to the problem at hand:
The consonants that I am going to talk about are these, listed in groups by similarity:
- ⟨ㅂ⟩, ⟨ㅃ⟩, and ⟨ㅍ⟩
- ⟨ㄷ⟩, ⟨ㄸ⟩, and ⟨ㅌ⟩
- ⟨ㅈ⟩, ⟨ㅉ⟩, and ⟨ㅊ⟩
- ⟨ㄱ⟩, ⟨ㄲ⟩, and ⟨ㅋ⟩
These sounds in Korean have various representations in romanization—particularly when no standard is being used and the person writing it is just going by ear, approximating it with the most similar English equivalent available. ⟨ㅂ⟩ can be ⟨p⟩ or ⟨b⟩, ⟨ㄲ⟩ may be ⟨kk⟩ or ⟨gg⟩, and ⟨ㄸ⟩ may be ⟨tt⟩ or ⟨dd⟩. But ⟨ㅍ⟩ is usually ⟨p⟩, and ⟨ㅌ⟩ is usually ⟨t⟩, and ⟨ㅋ⟩ is usually ⟨k⟩ with rarely any exception. Since the double consonants are usually unambiguously represented, the only confusion results from the representation of the single consonants—both in romanization and in the mind or the learner.
There is a good explanation for the confusion between the Korean consonant contrasts and the English consonant contrasts. The English brain distinguishes only voiced (vocal folds vibrate) and voiceless (vocal folds do not vibrate) pairs. The Korean brain distinguishes plain, tensed, and aspirated triplets. This results in some overlap when Korean romanization uses letters that may have other realizations in English.
English words themselves have many different kinds of pronunciation depending on context, many of which a native English speaker cannot passively perceive.
An English speaker may perceive 가 and 칼 as having the same sound (/k/) because they are both voiceless. But they are perceptually different to a Korean speaker. On the other hand, whether you say [kʰoŋ] or [gʱoŋ], a Korean speaker will hear /kʰoŋ/ (⟨ㅋ⟩).
Another unfortunate roadblock to learning the distinctions is also the fact that English loanwords in Korean tend to equate the plain consonants with voiced consonants and aspirated consonants with voiceless consonants. These examples may be familiar to you:
- Big Bang → 빅뱅 (/ɡ̚/ → /k̚/)
- cake → 케이크 (/k̚/ → /kʰ/)
- computer → 컴퓨터 (/t/ → /tʰ/)
This reinforces erroneous association between the sounds although I can see why the Koreans must do it—it maintains the contrast between the two English sounds in Korean even though the resulting sound change would be unnatural/robotic/hypercorrect in English (BIK-Bang, ca-KUH, compu-TUH). English only aspirates a voiceless initial consonant and voiceless initial consonant of a stressed syllable. The above examples would be better transliterated as 비그뱅, 케익, and 컴퓨더.
The solution really is to wean oneself off the English distinction between /k/ and /g/ and to train on the distinction between /k/, /k͈/, and /kʰ/—preferably by internalizing ⟨ㄱ⟩, ⟨ㄲ⟩, and ⟨ㅋ⟩. A change in perspective goes a long way to making sense of Korean phonology. I wish Duolingo had listening and speaking tests that contrasted these triplets.
Fantastic post, LiKenun. The differences of the k sound is one of my favorite parts of Korean phonology (can't say my username without using two of the three). These are great charts for illustrating the differences. From an English-speaking background, saying 'well, it's softer/harder than a normal k' is very vague and hard to grasp. I know on sites such as howtospeakkorean, the difference is made to be enormously difficult for non-native speakers to piece together, but knowing what to look for (and listen for!) makes the journey easier. Thoroughly appreciate this guide, I'll link it to one of my friends. :-]
I actually did not realize this way of looking at the consonant system until I heard some utterances in Korean which tricked me into thinking I was hearing another consonant. Then I realized that I was equating voicing with plainness, when the consonant was a voiced aspirate. The voiced aspirates don’t exist as a phonemes in English or Korean, but they exist in Hindi. It took a lot of listening to figure out. It’s like looking for a guy in a purple suit when the target should be a woman with horn-rimmed glasses. Many people listen for the wrong signals in the sound.
And so I wrote this as a resource to be linked to from other discussion threads since this kind of confusion is rampant. Feel free to spread the link! :)
I was discussing this with my friends earlier, how it's very difficult to determine what is changing between the three sounds. It does take a fair bit of practice in order to sort things out. Will definitely keep this in my back pocket. :-]
Do you have any more of those charts for different consonants? Very helpful!
Once you figure out the speech mechanisms for one set of consonants, you pretty much don’t need examples for the other ones. You just substitute /g/ with /d/ and /k/ with /t/ and find some words that contain those sounds…
The general pattern for English is that fortis consonants become aspirated at the beginning of words and when stressed. They are tensed before or after another consonant like s, p, t, or k. Example: diaper will give you the pronunciation for ㅂ, spit will give you the pronunciation for ㅃ, and panther will give you the pronunciation for ㅍ.
Thank you for the ㅂ,ㅃ,ㅍ example. The reason I asked is there are some consonants that I don't know if there there are proper examples for.
For example, jar (ㅈ), char (ㅊ), and jaw (ㅉ) seem to work for me, but how about the difference between ㅅ and ㅆ, or ㄷ and ㅌ? Those are the ones in particular that I don't hear a huge difference for, or at least I can't think of any analogies for English maybe because one doesn't exist. Do you have any advice for those consonants specifically?
For the j/ch series:
- ㅈ: handkerchief
- ㅉ: mischief
- ㅊ: chief
ㅅ and ㅆ are tricky to differentiate with English approximations. The English s is closer to ㅆ than ㅅ, so basically, you already know what ㅆ sounds like. ㅅ is softer sound which I can only describe as a loosening of constriction of airflow from the throat to the lips.
ㄷ, ㄸ, and ㅌ are also tricky, particularly because intervocalic t in English tends to weaken to [ɾ] in many dialects, which would be a Korean ㄹ. Aspirated t also tends to exhibit a fricative release similar to ts in cats. Ignoring those oddities, the pattern would be similar to that of the other series:
- ㄷ: avatar
- ㄸ: star
- ㅌ: tar