Using {da'i} with tru= th-functional connectives is, of course, dissonant, since a {da'i} sentence= has no truth value. But w= e do allow this is some other cases, so perhaps this is no harm. Whil= e it is true that "I suppose" and the like don't generate conditionals in a= direct way, the usual reasons for such suppositions is to consider what wo= uld happen if the supposition were true. This is a longer task than c= an be handled in a simple conditional, although the results can be summariz= ed in one -- and usually is. Hence the paragraph marker suggestion (o= n analogy with the inset or boxing conventions of ordinary logic for reductio and conditional proof).

From: And Rosta &l= t;and.rosta@gmail.com>
To: lojban@googlegroups.com
= Sent: Friday, August 31, 2012 2:39 PM
Subject: Re: [lojban] Revising mu'ei and CAhA on= ce again. Possible worlds.

la gleki, On 31/08/2012 17:48:
> On Tuesday, August 14, 2012 10:34:20= PM UTC+4, And Rosta wrote:
>
> Gleki Arxokuna, O= n 14/08/2012 18:23:
> > I wish Robin started u= sing {mu'ei} again but
> > it's really when us= age decides. May be human brain just doesn't want
> &nbs= p; > to deal with A-level at such level of precision. May be {ka'e/na> > ka'e/ka'ei/bia'i} or even {bi'ai} is enough.=
>
> The evidence of natural language is to the c= ontrary. The could/probably/would contrast is the some/most/all contrast.>
> > The use of {da'i} is interesting. F= or a logical language it's
> > completely depl= orable, because there's a complete mismatch between
> &n= bsp; > the lexicosyntactic form and the logical form, and no explicit rule
> > about how to get from one t= o the other -- it works by mere stipulated
> >= magic. But it caught on among those impatient to be actively using
>= > the language, and nicely illustrated the fundamen= tal incompatibility
> > between a loglang and = a language governed by the principle of "let
> &g= t; usage decide".
> >
> &nb= sp; > May be we can determine the most common usage of {da'i} and redefi= ne
> > it from the point of view of A/M/F-leve= l scheme? May be we should
> > perform analysi= s of Lojban corpus and tatoeba sentences?
>
> {da= 'i} is in UI, isn't it? So it doesn't have the right grammatical properties= .
>
> It is in UI. If I "discovered" A and F levels why not bind {da'i} to A-level i.e. make it a synonym of {ka'e} but without ch= anging the grammar and selmaho
> and {da'inai} would be "equal" to {c= a'a}.

Because {da'i} should be a marker of mood -- of hypothetical, = unassertive mood; whereas, ka'e is a modal of possibility. Modals involve q= uantification over possible states of affairs of various sorts. Moods invol= ve a relation between the speaker and the proposition -- the speaker assert= s p to be true, the speaker wishes p were true, the speaker entertains the = idea of p, the speaker asks whether p is true, and so forth.

(There'= s no harm in marking the protasis and/or apodosis of a conditional with da'= i, but da'i doesn't generate conditional semantics.)

--And.

-= -
You received this message because you are subscribed to the Google Gr= oups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To= unsubscribe from this group, send email to lojban+uns= ubscribe@googlegroups.com.
For more options, visit this group at htt= p://groups.google.com/group/lojban?hl=3Den.