* Tuesday, 2014-11-04 at 20:23 -0300 - Jorge Llambías <jjllambias@gmail.com>:
> On Mon, Nov 3, 2014 at 10:09 PM, Martin Bays <mbays@sdf.org> wrote:
>
> > So then a tag would be able to play three roles:
> > (i) combining with an optional term to form a modal operator;
> > (ii) acting in certain connectives as a binary relation on events;
> > (iii) acting in the {pe [tag]} construction as a binary relation.
>
> Shouldn't (ii) be split into (ii-a) a tag acting itself as a connective,
> and (ii-b) a tag modifying some other connective. It seems that not all
> tags can do (ii-a), only tags that can take two event/proposition arguments
> can do do that, whereas in principle all tags could do (ii-b), depending on
> how the modification of the base connective goes. Also (ii-a) may, but
> hopefully doesn't, split into two further cases: "[tag] gi ... gi ..." and
> "... .i [tag] bo ..."
>
> > It would be nice to unify (ii) and (iii).
>
> Doesn't (iii) always reduce to (i) though: pe [tag] ko'a = poi ke'a co'e
> [tag] ko'a?
CLL has it being less vague, though; e.g. {ko'a pe cu'u ko'e} is meant
to really be equivalent to {ko'a poi ko'e cusku ke'a}.
> > I don't see a natural way to fully reduce (ii) to (i) or (i) to (ii).
>
> I think when the tag tags an event as sumti tcita. the relationship between
> (i) and (ii-a) is pretty straightforward:
>
> broda .i [tag] bo brode ~ broda [tag] lo nu brode (for non-tenses)
>
> broda .i [tag] bo brode ~ broda .i brode [tag] lo nu broda (for tenses)
Concentrating on this tense case:
I don't really see how this {lo nu} would end up working.
e.g. in CLL's {mi klama le zarci .i ba bo mi klama le zdani}, and
supposing the context is such that there's a specific event of
market-going involved, presumably that forbids a kind reading of {lo nu
mi klama le zarci}, and so it will have to be a plural - the
(contextually relevant) events of my going to the market, of which there
may be many spread over a large swathe of time. Am I then meant to be
claiming that my going home is after all such events? Or at least one?
Or "most" of them?
I would say that the seltcita sumti should specifically involve the
event(s) involved in the first connectand.
> With forethought I would drop the independent "broda" that appears in the
> tense case.
>
> (ii-b) is the tricky case, because the logical connective wants to evaluate
> the truth values of two propositions independently, so in which of the two
> propositions does the tag play a role?
>
> > > The mix with logical connectives is the tricky case, but it becomes easier
> > > if we think of the tag as modifying only the second connectand, in which
> > > case "broda .i je tag bo brode" is just (broda) .ije (tag(brode)).
> >
> > For tenses, yes.
>
>
> And for non-tenses, by analogy I think it has to be
>
> broda .i [tag] bo brode ~ broda .i broda [tag] lo nu brode
Did you mean to have a {je} here, and for it to be different from the
expansion of {broda .i [tag] bo brode} you gave above?
> > The only tricky bit is deciding what exactly the seltcita sumti of
> > the tag should be.
>
> "lo nu xu kau broda"., "the event of brodaing or not brodaing, whichever
> the case might be".
But which event(s)?
> > OK. So it currently seems to me that:
> >
> > * quantifying over events is the right thing to do;
> >
>
> If by that you mean the speaker is claiming a hidden "su'o", I have to
> disagree.
>
> Going back a bit, when someone says "carvi" they are not saying "there is
> some event of raining, x, and x happens". They may be saying "c is an event
> of raining and c happens", but not the first. If they were saying the
> first, there would be no way to negate the claim, because "na carvi" would
> be "there is some event of not raining, x, and x happens", which doesn't
> contradict the first claim, so "na carvi" would not be the way to negate
> "carvi".
But in a specific world at a specific time and position, it rains xor
not-rains, and in either case there's an event of raining / not-raining
enveloping that particular point. So "some event of not raining happens"
*is* equivalent, pointwise, to "no event of raining happens".
In other words, we're meant to be within the scope of any quantifiers
over spacetime/worlds when we make the substitution {broda} <-> {su'o nu
broda cu fasnu}.
I don't really know what {carvi} means - it depends on how exactly one is
meant to interpret this idea of "implicit tenses" - but I think that by
this argument it must be the same as whatever it is that {su'o nu carvi
cu fasnu} means.
Usually this would be a pretty empty move to make, but I think it helps
with these connectives. {broda .i ba bo brode} <-> {da zo'u da nu broda
kei gi'e fasnu .i bo ba da brode}.
So e.g.
ca ro nu mi xagji kei mi klama lo zarci .e ba bo lo zdani
-> ca ro nu mi xagji kei da poi nu mi klama lo zarci zo'u ge da
fasnu gi ba da mi klama lo zdani
Even if that isn't the best way to formalise it, I feel sure that
something along these lines must be the right meaning, and that
ca ro nu mi xagji kei ge mi klama lo zarci gi ba lo nu mi klama lo
zarci kei mi klama lo zdani
is missing something crucial.
> Now, when A says "carvi", we may describe this as "there's some event of
> raining, x, and A says that x happens", but that's not the same as saying
> "A says that there's some event of raining, x, and x happens".
> So if by "quantifying over events" you mean that the speaker is quantifying
> over events, I don't see it.
>
> If you mean that you can interpret what the speaker is saying by
> quantifying over events, (i.e. the metalinguistic interpreter does the
> quantifying, not the speaker) then that may be,
I'm not sure I'm exactly saying either of these. I'm saying that {carvi}
is equivalent to {su'o nu carvi cu fasnu} in much the same way that it's
equivalent to {na na carvi} (although with a rather more complicated
logic involved). So what A says has the meaning of a quantification over
events.
> but isn't the parser supposed to just translate from one language to
> another rather than make a metalinguistic interpretation?
Yes.
> > * for tenses: the second sentence gets tagged, with the seltcita sumti
> > being (the variable for) the event of the first sentence, which is
> > declared to occur (unless the connective is forethought and below
> > sentence level);
> > * for non-tenses: the same, but with the roles of the first and second
> > sentences swapped;
>
> We're probably not disagreeing very much about those two, although we may
> quibble about a detail or two.
I don't particularly like all these details; I'm trying to follow CLL
even where it seems strange.
> > * when a logical connective other than {je} is involved: we have to
> > separately consider the cases that the seltcita sumti involved is an
> > event of the sentence and that it's an event of the sentence's
> > negation.
>
> ... because in order to eveluate a logical connective you have to consider
> cases where the proposition is not true as well as cases when it is true.
Quite so.
Martin
Attachment:
signature.asc
Description: Digital signature