[Japanese text in email body]

It is definitely deletion (though devoicing may have indeed been a prior
stage and is still discernible across word/phrase boundaries), and it does
not occur only with /M/, only that it is the most common vowel to be
deleted. A look at rapid/colloquial speech easily demonstrates:

The polite "するのです suru no desu" /sMrMnodesM/ gets realised [sMrnodes] in
rapid speech, and in colloquial form is [sMnnodes]. In turn it is
represented in e.g. a conversation in a novel or a speech bubble in a manga
as "すんのです sunno desu".

A more telling example is "わからないのです wakaranai no des(u)", which goes through
the various carefulness/speed stages:

"わからないんです wakaranain des(u)"
"わからないんっす wakaranainssu"
"わかんないっす wakannaissu"

…and among close friends decays simply into "わかんない wakannai". If the emotive
particle "よ yo" is appended, all the above forms ending in "-su" drop the
vowel and are pronounced with [sjo].

Other examples include "洗濯機 sentakuki" [sentakki] (washing machine), "首になる
kubi ni naru" (to get fired) where the "-i" in "kubi" is one of the examples
of trans-word devoiced vowels I mentioned above.

Interestingly, the word "生きる ikiru" (to live) /i'kirM/* when inflected for
the progressive tense "生きている ikite iru" changes prosody to /'ikite.irM/ and
is realised as ['iCteirM].

It is also not uncommon to hear, especially in male speech, "人 hito"
realised as [Cto]. But because Japanese is a moraic language, no
syllabification issues occur.


* Using the stress mark to denote the syllable with the high pitch (at least
in the standard Tokyo dialect)

2010/2/25 Chris Peters <[log in to unmask]>

> > On Thu, Feb 25, 2010 at 2:53 PM, Lars Finsen <[log in to unmask]>
> wrote:
> >
> > > Having just watched a film about the great Hokusai, it strikes me that
> his
> > > name is pronounced something like /hoksai/. No trace of the middle
> vowel
> > > there to be heard, not even the slightest remnant of a schwa. Is this
> > > phenomenon prevalent in other Japanese words, and do we have any
> information
> > > of the histories of such vowels? I believe I have heard something
> similar
> > > happening with final vowels in words like wakarimasu. Is it always the
> u?
> > > U's in other positions, like Ueda or Sugimori, seem to survive.
> > >
> As a partially-fluent Japanese speaker, here's the rule of thumb I've
> always gone by.  The vowels "i" and "u" are diminished (or sometimes dropped
> entirely) if they occur between voiceless consonants (i.e. "Hokusai"), or at
> the end of a word when they follow a voiceless consonant ("wakarimasu").
> _________________________________________________________________
> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.