Hello,
I’m am very new to Japanese. Are there rules governing when certain value sounds are not pronounced?
For example why is 少し (すこし) (a little) pronounced “skoshi” and not “sukoshi”. And why is けしき (a view) pronounced “keski” and not “keshiki”?
Thanks,
David
This is the thing:
Since you can hear it I wouldn’t worry about it too much. It’s a natural thing that happens in most languages.
Just pay attention to when it happens and you’ll pick it up.
To be fair, vowel devoicing is not actually very common in languages.
What’s is your bar for common? It happens to some extent in languages throughout Europe and Asia, independent of language family, and for instance Native American languages also do it.
It’s just a natural part of talking fast (can’t speak for tonal languages though).
It happens in Canadian French for sure. For example: couteau, titan, citation, super and chicane when spoken at a normal speed often have their first vowel devoiced. Of course the vowels have to be close vowels.
I never noticed it before learning Japanese.
Old English used to have devoicing as well. I’d say it’s a pretty normal consequence of sound production given the equipment we use.
Devoicing happens all the time in latin derived languages. Spanish, Portuguese and Italian, for example.
Happens in German as well. Probably why it’s got a reputation for being a language that consists mostly of consonants. 
To be clear, I’m talking about vowel devoicing and not vowels becoming silent, those are not the same thing. When vowels become silent in English or German, those vowels are not pronounced at all, however in Japanese the vowels are pronounced as a breathy aspiration. That kind of sound doesn’t exist in English. I’m not a linguist so perhaps I should refrain from making claims about whether something is common or not.
“Most languages only have voiced vowels.”
Hello, Linguist here.
Vowel devoicing might be more common than you think. It happens in quiet a few languages (Including English and German); however, it is commonly treated as allophonic variation.
I’d say the difference is in the fact that Japanese is mora-timed rather than syllable-timed as far as isochrony. The devoicing is the same process, it’s just more apparent in Japanese in order to account for the duration of the devoiced mora, but I think devoicing itself is present in almost all languages to some degree.
Yes, that’s, uh, exactly what I was trying to say. 
Just tested with a microphone and I definitely devoice vowels in English when talking fast (although the vowel usually becomes a schwa first).
I guess it might be worth pointing out that when you’re devoicing, you move your mouth as of you were going to pronounce the vowel anyway (unlike you would in English).
I’d be interested in a few examples of where it happens in German. And to know where that reputation comes from.
I’m not a linguist, but an Austrian 
I believe it happens all the time, for example:
“sagen” → “sag’n” (to say)
a more extreme example:
“gewickelten” → “g’wick’lt’n” (past participle of “to wrap”)
From my observation, different dialects do this differently. In Austria, many dialects shorten many of these vowels to the point of omission. In the first example above, no one ever says the full “e”, except maybe for emphasis in a stage drama or so. In many places the contraction even goes further and turns the “g’n” to a “ng” sound. In the second example, the first “e” would not be shortened in formal speech, but the second “e” would be.
Please correct me if this is actually a different situation! It certainly seems similar.
Thanks for that. I’m an English native speaker, but I’ve lived and worked in Germany in and off for the last 40 years, and have a background in linguistics.
I’m naturally familiar with that sort of vowel reduction, especially in unstressed syllables in rapid speech, and certainly this feature is even more evident when ‘Mundart’ is spoken. Indeed, in Rheinland Pfalz, where I live, consonants are also frequently lost.
Japanese is a little different to this, I think, in that i and u are in many cases lost, or rather, not spoken at all, even in formal speech, and during my experience of living in Japan for some years it seemed to me that the final vowel in です for instance was never pronounced except as a sort of conscious flourish.
I think it’s still usually noticeably pronounced, but shortened and devoiced. Difficult to describe, but in English if you were saying “dress” your mouth stays in the position of the preceding vowel (/ɛ/) sound, but for です your mouth ends in the /ɯ/ position from す, even if you don’t explicitly “say” the vowel.
I stand corrected, and humbly abase myself.
Thank you everyone for well thought and the scholarly responses. Very helpful. I have a lot of new things to read up on.
That’s a bit awkward…I never know what to do when people humbly abase themselves
Just wave your hand back and forth with your face like 
and keep saying stuff like いいえいいえ、とんでもないんです while bowing repeatedly. You get used to this in Japanese culture