Ref Request: Adverbial Scope in Japanese

[Ryan Ginstrom]

Thank you very much for your response. As I feared, it seems that this
matter has not been explored in the literature. Perhaps this is because
there is nothing of interest here. Nevertheless, allow me to respond
to your points below.

On Sun, 23 May 1999, GUNJI Takao wrote:
> ryan> (1) waza-to 'on purpose' vs. waza-to-rashiku 'seemingly on purpose'
>
> ryan>    a. Kaoru-ga  Masumi-ni  waza-to    korob-ase-ta
> ryan>       Kaoru-NOM Masumi-LOC on.purpose fall-cause-PAST
> ryan>       'Kaoru made Masumi fall on purpose'

	Please excuse this error. The sentence used with my informants
contained 'Masumi-o'; the use of the dative marker here was a non-native
error on my part.

>  Kaoru-ga Masumi-wo korob-ase-ta.

	I presume we can agree that inside scope is at least
difficult for this sentence (three native speakers have told me it is
impossible, and you say your intuition is unclear). Note that no one seems
to have any problem with the evidential waza-to-rashiku getting inside
scope.

> Thus, a more decisive evidence, if any, would have to involve either the
> accusative marker or a volitional verb.
>
> Kaoru-ga Masumi-wo wazato korob-ase-ta.
>
> Kaoru-ga Masumi-ni wazato sippais-ase-ta.
>                           fail

	I have only been able to gather one native intuition for this
sentence so far, but this informant can only get the outside scope
reading for this sentence as well. Together with your unclear intuition, I
would like to provisionally give the inside scope reading a question mark
as well. Again, my informant did not have any trouble getting inside or
outside scope with 'waza-to-rasiku,' nor did you.

	Interestingly, my informant noted that for the folowing sentence,
either Kaoru or Masumi could have been waza-to:

Kaoru-ga Masumi-ni waza-to shippai s-ase-rare-ta
                                         PASSIVE
'Kaoru was made to fail by Masumi on purpose'

This seems to (1) reinforce the fact that subject-orientation can be
either syntactic or semantic, and (2) show that in the causative sentence,
the dative complement is not accessible to 'waza-to' as either a semantic
or syntactic subject. Again, I feel this points to a monoclausal analysis
of causative constructions.

> ryan>    a. Kaoru-ga  Masumi-ni  heiki-ni tabe-sase-ta
> ryan>       Kaoru-NOM Masumi-LOC calmly   eat-cause-PAST
> ryan>       'Kaoru made Masumi eat calmly'
>
> This is simply ill-formed.  If you want to express the meaning of
> 'calmly', something like 'sizuka-ni' would be more appropriate.

	I am not sure what you mean here. Is the sentence ill-formed,
or do you not like my gloss of 'calmly?' If the latter, how about
'cooly?' At least some native speakers find this sentence well-formed, and
agree that only Kaoru could have acted 'heiki-ni.' Perhaps it would be
more acceptable to you if you imagined Kaoru making Masumi eat a worm.

> Kaoru-ga Masumi-ni sizuka-ni tabe-sase-ta.
>
> In this case, the interpretation in which Masumi acted calmly seems
> preferable, despite your judgment.

	Note that shizuka-ni is neither a speaker-oriented nor
subject-oriented adverb, which I am collapsing into the category
"point-of-view adverb." Thus I do not feel it has any bearing on my claim
regarding point-of-view adverbs and scope.

> If you have 'heiki-de' instead of 'heiki-ni', you get a well-formed
> sentence:
>
> Kaoru-ga Masumi-ni heiki-de tabe-sase-ta.
>
> This seems to have only the 'Kaoru acted calmly' interpretation, but
> you cannot say 'heiki-de tabe-ru' in the first place. So, this seems
> to be related to some kind of semantic mismatch and may not be related
> to scope.

	Out of curiousity, do you allow either of the following?

Kaoru-ga  heiki-de/ni mushi-o tabe-ta
Kaoru-NOM cooly       bug-ACC eat-PAST
'Kaoru cooly ate a bug'

Kaoru-ga  heiki-de/ni hito-no    ranchi-o tabe-ta
Kaoru-NOM cooly       person-GEN luch-ACC eat-PAST
'Kaoru cooly ate someone else's lunch.'

	So far the one native speaker I have polled finds nothing wrong
with these sentences (she suggested the second one herself).

	I'll assume you do not allow 'heiki-de taberu' or 'heiki-ni
tabe[-sase]-ru,' but let me reiterate that at least some native speakers
judge these to be grammatical, and their judgements w.r.t scope support my
claim. Also note that I have consciously avoided ADVERB-de type modifiers
such as heiki-de, as these presumably have a null argument which could be
coindexed with either the actor or undergoer, thus clouding scope
judgements.


Regards,

Ryan Ginstrom
email:  ryan at csli.stanford.edu
        ginstrom at ling.ohio-state.edu
URL: ling.ohio-state.edu/~ginstrom/