[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: imperatives



> >         la kim cu cismyfra do e'o
> >         That Kim smile at you, I ask you.
> >
> > That is still something strange to say, since the agent here is Kim,
> > so it is strange to ask you that she smile. It's like saying "be smiled
> > at by Kim".
>
> Why? I find it perfectly valid in lojban.

I find it perfectly valid too, but still strange.

> To get at the 'agent' you have
> to reach into deep case theory, and I don't remember anybody relating
> lojban with Fillmore.

I meant it in the intuitive sense. :)

> I don't find di'e chain of reasoning unreasonable, pei?
>
> I ask you that Kim smiles at you.
> I ask you that it happens that Kim smiles at you.
> I ask you to do the best you can for Kim to smile at you.

This is the critical link. First you were asking me that Kim do something,
and now you are asking me to do something myself, so that something else
happens.

Asking me that Kim smile at me is like asking me that Kim sneeze.
The fact that {do} is present as an argument in one case and not in the
other is accidental. In both cases, I coould take some action to
bring about the result, but that is not what you are asking in the
first case, at least not directly.

Compare:

        doi goran e'o le plise cu cpana do
        I ask you that the apple be on you.

        doi goran e'o le plise cu cpana le jubme
        I ask you that the apple be on the table.

They are exactly equivalent sentences. The fact that the first one
has {do} as one of the arguments is accidental. Those who like using
{ko} in non-agentive places might want to say {le plise cu cpana ko},
which is perfectly valid Lojban, but it is just a cheap trick, there
is no semantic basis to distinguish the two cases of {do} and {le jubme}.

> ta'o did you read The Little Prince (is that correct?) by Antoine de
> Saint Exupery (sp?)

That's the spelling, bar some accent or other.

> There was a conversation between a king and the main
> character that looked more or less like this (I can't find the book right
> now, I'm doing this by memory):
>
> .i le cmanolbe'a cu senci
>
> .i cu'u le nolraitru lu
>    .i .o'onai mi punai curmi lenu do senpi li'u
> .i cu'u le cmanolbe'a lu
>    .i lenu mi senpi cu selbai gi'eseki'ubo sarcu li'u
> .i cu'u le nolraitru lu
>    .i lenu do senpi cu sarcu xu .i .i'a seni'ibo .e'oga'i do senpi li'u
> .i cu'u le cmanolbe'a lu
>    .i ku'i mi ca na kakne lenu senpi li'u
> .i cu'u le nolraitru lu
>    .i .o'onairo'a ku'i do puzi senpi .ije mi minde do lenu senpi li'u
> .i cu'u le cmanolbe'a lu
>    .i mi senpi ri'a le da'i nu sarcu kei .enai lenu mi djica li'u
> .i cu'u le nolraitru lu
>    .i .ua .i ni'i la'e di'u fau lenu lenu senpi cu sarcu kei ko senpi
>    .ije fau lenu lenu senpi na sarcu kei ko na senpi li'u
>
> Do you really think that in "mi senpi", I am the actor?

You changed from {senci} to {senpi} after the first time. In both
cases it is doubtful whether you are the actor or not, which only
strengthens my case. :)

> pe'i I am a
> patient. That doesn't stop the king from ordering me not to sneeze, and
> it makes just as much sense as ordering somebody to be smiled at.

And it makes as much sense as ordering me that pigs fly. My point is
that the one being ordered need not always be one of the sumti of the
desired relationship.

> That
> is not under sy. control, or at least not completely, but there is
> nothing impossible in the request itself, even if its fulfillment
> logically can not really be expected.

I don't dispute any of what you say. I still think that commands to
someone other than the agent of an action are strange, even if
perfectly valid and possible.

Also, I don't think there is any reason other than perhaps frequency
of use to single out {do} to have an imperative version of its own.

Jorge