Normally I try to stay away from writing about pronunciation too much on a text blog – after all using words you can only say so much about how things sound. Some things are best learned in person, or at least with an audio blog or podcast (which I may try to do someday).
But I was recently inspired by 1994sunshine’s blog post about Japanese pronunciation, so I thought I would write a post with some hints on the same.
The gist of 1994sunshine’s post is that Japanese pronunciation isn’t as simple as just sounding out the sound from each hiragana letter in order to make words. There is some changes in how things are said depending on the word. The example cited in that post, which is a good simple one, is the word “すき” which sounds closer to English ‘ski’. I wanted to add a few thoughts to this discussion, specifically about Japanese pronunciation in general, and also about a few other sounds that get transformed like the す sound in this example.
In Japanese, things are typically pronounced very evenly with a regular tempo that is a good bit faster than that of English sounds, on average. One way you can see that is by looking at many loan words which get extended with a “-” to make them sound closer to their English counter parts. Take for example, “database”, which turns into データベース, or “safe” which can be written as セーフ. Sometimes there is also a pause inserted (via a small つ) as in the case of “happy” (ハッピー). Translating these words straight to katakana without adding a pause or lengthening would make them just too quick to sound like the original English.
As for the regular tempo, remember that Japanese doesn’t have a stress accent like English, where there is emphasis on one syllable (like “re-MEM-ber”). Rather the pitch changes throughout the word and sentence, sometimes suddenly and sometimes more gradually. See my post here for a more detailed explanation. Of course certain syllables can be accented for emotion at the speaker’s desire, but that is different from the accepted way of pronouncing each word.
The lack of stress accent is good to keep in mind, since beginners to Japanese may at first try to accent certain syllables, which leads to unnatural-sounding words. Instead, try to say everything in a metronomic, even fashion, and keep things light.
Now that you understand the general feeling of how Japanese is pronounced, it may make a little more sense to you why words like “すき” end up sounding like “s-ki”. Sure, you could pronounce the “す” fully, but omitting it allows a more even, metered pronunciation of the word. Keep in mind that saying “suki” isn’t technically wrong, and in rare cases I have heard natives say it that way (usually women), but the average person is going to say things the most efficient and smooth way possible. Also when saying letters on their own, like when saying the hiragana alphabet, you always sound things out completely, so す always sounds like “su”.
Now let’s see a few other examples where sounds get transformed when in the context of a word, and said at a normal speed in conversation.
- In 期間 (きかん、time period), the ‘i’ sound fades away into a breathy wind.
- In 試食 (ししょく, food tasting), the ‘i’ sound in the first ‘shi’ is very de-emphasized.
- In 人（ひと、person), the ‘i’ sound in ‘hi’ is said very lightly or not at all, and sometimes this word ends up sounding like “sh-to”.
- In 飯（めし、food), the “i” sound i “shi” often gets omitted, resulting in “me-sh”.
- In 宿題 （しゅくだい、homework), the “u” sound in “shu” often gets omitted, leaving “sh-kudai”.
- In 大学 （だいがく, college), the “u” sound in “ku” sometimes gets omitted, as “daigak”.
I think you’re starting to see a pattern emerge. Many times sounds in the ‘i’ row (き／し／ち／。。。） or ‘u’ row (く／す／つ／。。。), when followed by another sound (except for pure vowels like あ／い／う／え／お), end up loosing part or all the ‘u’ or ‘i’ sound. This can also happen when it is at the end of the word (like with めし or だいがく).
I’m not going to go so far as to call this a hard-and-fast rule, since it depends on the word, but once you understand these general patterns you’ll find it easier to pick up pronunciations of words, and have a better chance of guessing the right way of saying things the first time, even if you’ve never heard a word spoken before.
Another pattern that is good to discuss here is when a word’s pronunciation is simplified by getting rid of a syllable and replacing it with a small つ。For example,
- 洗濯機 (washing machine): せん + たく + き ＝＞ せんたっき
- 国家 (country, state): こく＋か ＝＞ こっか
- 血管 (blood vessel): けつ＋かん ＝＞ けっかん
This fits right in with the Japanese way of optimizing things to make them easier, and faster to pronounce.