Re: Words that are just speech acts

On 10/29/06, And Rosta <and.rosta@...> wrote:
> > To a certain extent NLF2DWS is a way of writing (diagramming?) ideas.
> > Performatives or pure "speech acts" would thus probably need to be
> > treated somewhat specially IMO, since there is not necessarily an
> > assumed context that the text is merely transcribing actual speech, so
> > the meta (like this) should be treated explicitly.
>
> I don't see why NLF2DWS should be different from ordinary one-dimensional
> speech in this respect. Both are ways of expressing ideas. In both systems
> I'd expect the basic unit of what is expressed to be the speech act (which
> may have propositional content).
>
> How do you envisage NLF2DWS handling imperatives and interrogatives?
I'm not really sure to be honest.

It seems it would best work for handling more abstract things - for
simply laying out a thoughtweb, or a story, or an idea itself, rather
than a *speech act* per se like an imperative or interrogative.

This relates to (or is perhaps just a reiteration of) e.g. the
question of "how would you tell a story?". 'cause I don't think you
*could* really do it in at all the same manner as you would in a
normal language; it wouldn't be psuedoconversational. Or at least, if
it is, it'd be a mode of conversation that we don't yet have a good
model for. Whereas any linear language can base itself on that pretty
fundamental set of assumptions about how conversation (and oratory
etc) works.

I'm not sure how you would do such things when there *isn't* a clear
conversation-model that is being recorded or based upon. You could
mark things as being desired or required by someone, but that doesn't
have the same force as an imperative. You could likewise mark things
as being unknown or "fill in the blank here", but again without a
conversation you have no reason for the reader to feel that they are
obliged to respond (or simulate a response).

 - Sai