Re: ReTonogenesis

>From: Henrik Theiling <theiling@...>
>
>Kevin Athey <kevindeanathey@...> writes:
> > said to fall on mora breaks.  BUT, having studied a lot of African
>tonality
> > in school, my interpretation would be the other possibility:  that of
>pitch
> > spreading, in which the accent is really H, which spreads left, with an
> > additional rule to drop initial unaccented morae to L.
>
>With this interpretation, how do you get three possible accent
>patterns for a *bi*moraic word, when the accent *on* the morae instead
>of in between?  I did not get this, I think.
>
>BTW, I left out the detail of actual pitch change since a) I did not
>remember them :-), b) they seem to be very different from dialect to
>dialect, so the normal description holds for Tokyo accent only.  (At
>least I remember a large table in 'The Languages of Japan').
If there _is_ a Japanese expert out there, he or she should jump right in.

OK, if I understand you properly, you are asking how a sequence of phonemes
consisting of two morae can have three possible tonal contours.

When you are dealing with the "word" standing alone, it doesn't.  It can
only have two:  LH or HL in the Tokyo dialect.  (The Kyoto dialect, I
believe, allows HH and HL, and the words are generally switched, strangely
enough.)  However, if you were to add the topicalizing particle to this
"word", the ambiguity would be lost.  I don't have my dictionary marking
accent with me, so I can't cite specific words, but let us use /hashi/,
which is bimoraic.

'ha.shi => HL
'ha.shi wa => HLL
ha.'shi = > LH
ha.'shi wa => LHL
ha.shi => LH
ha.shi wa => LHH

The latter two are ambiguous in the citation form, but rarely in a sentance.

There is the added complication that the actual domain of pitch accent in
Japanese is the phrase rather than the word, but that's basicly my
understanding.  I hope that's clear.

Athey

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Replies