Date: Tue, 14 Feb 1995 23:25:18 +0000
From: ucleaar <ucleaar@UCL.AC.UK>
Subject: Re: ago24 & replies
In-reply-to: (Your message of Mon, 13 Feb 95 22:23:07 EST.)
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Reply-to: ucleaar <ucleaar@UCL.AC.UK>
Message-id: <01HN40VXK9WG000OW3@FIPORT.FUNET.FI>
Content-transfer-encoding: 7BIT
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Length: 5855
Lines: 130

Jorge:
> But can't an object be important then?
Well the notion of importance has an element of conditionality in it:
"It is important that she pass the exam" means, roughly, "If she passes
the exam then things will be better and if she doesn't then things will
be worse". But there's no reason not to have a selbri glossable as
"important" that takes an object as x1, with the selbri defined as
"if the bridi implied by x1 is the case, then ... etc.".
I've said all this before re nitcu. We can have all the selbri we
want, but there's only so much work a given selbri can do.
 
> > > Even if we can't agree on nitcu specifically. Do you think that there
> > > is (or should be) certain tersumti that accept _exclusively_ irrealis
> > > events?
> > I think there should be certain tersumti that accept exclusively
> > irrealis sumti: i.e. le+anything,
> I hope you are not saying that irrealissness and nonveridicality are
> the same thing!
 
No, I'm saying that nonveridicality allows the sumti to refer to anything
at all, including nonreal things not found in this world.
 
> > or {lo bridi}
> Isn't that an object?
 
I'm not clear what an object is. At any rate, surely {lo bridi} refers
to a bridi, just as {lo duhu} does.
 
> Can you say {mi nitcu lo bridi}, meaning that
> you need that what certain bridi says be true?
 
I reckon so, if {nitcu} is defined as "it is advantageous to x1 that
x2 be the case".
 
> > I don't have any intuitions on the relationship between (ir)realis and
> > genericity. In a sense I think generics are purely mental objects, and
> > so are neither realis nor irrealis.
> Aren't irrealis events purely mental objects as well?
 
In the sense that they have no extramental existence in this universe,
but not in the sense that they are conceptualized as concepts. You
may not share my feeling that we conceptualize {lohe broda} as a
concept.
 
> > but properties ascribed
> > to lohe mlatu as a sumti are understood to be properties not of the
> > conceptual archetype but of instances of the archetype.
> I don't agree with this, but then my idea of {lo'e} is not the standard
> one. I don't think that any instance of the archetype need exhibit the
> property. A claim about the archetype is a claim only about the archetype
> and the other objects that are claimed to be in a relationship with it.
 
I see generics as analogous to (intensionally defined) sets: the
set has properties and the members have common properties. It simply
has to be stipulated whether {lohe broda cu brode} means the "set"
is a brode (your preference), or whether it means the "members" are
brode (my preference, and the present norm). Ideally we'd have
distinct ways to do both (maybe using something from LAhE).
 
> To me {mi djica lo'e nu mi klama} means that {mi} and {lo'e nu mi klama}
> are in relationship {djica}, so I am in relationship {djica} with the
> archetype of events of I go. There is no claim about any particular event
> of me going.
 
That's okay under the way you use {lohe}. But it's not clear to me that
it means "I want to go". It's analogous to saying "I want a set of
goings" - what would that mean? What sort of properties does an archetype
(as opposed to its instances) have?
 
> If only propositions can be irrealis, what do
> you call the event described by an irrealis proposition?
 
Well I changed my mind since I wrote that, persuaded by "I described
a [nonreal] book". I therefore won't try to answer your question.
 
> > Also, to me (and nothing from LLG leads me to
> > think I'm right on this) {lohe} is like a default universal quantifier.
> To me it is nothing of the sort (nothing from LLG leads me to
> believe that I'm right either). If you want {ro} or {so'e}, why would
> you use {lo'e}?
 
Because neither {ro} nor {sohe} is a default universal quantifier: they
do not involve typicality.
 
> > So if {lohe mlatu cu xekri} then a white cat is an exceptional, atypical
> > cat.
> Why wouldn't you just say {so'e mlatu cu xekri} for that?
 
One reason is that this is a statement of objective fact: you can test
it by inspecting the extramental world. In contrast, to test {lohe mlatu
cu xekri} you have to examine our mental representations of the world,
to examine the cat archetype & see whether blackness is deemed to be
a property of typical instances of the archetype.
 
> And what would it mean that the event is atypical?
 
that it is unpredictable, mildly surprising, marked, with properties one
wouldn't have expected it to have. There is a large literature on
typicality in prototype theory.
 
> I'm quite happy with {le} for uniques. In most cases it labels
> a unique thing (the one you have in mind), and even when it doesn't
> label one single thing, it labels many things taken individually.
 
The drawback is that it is nonveridical. I know you don't approve
of that, but thus spake LLG.
 
> > In fact, I rather fear that might not work. Consider {troci}:
> > {koha pu troci le duhu koha klama} (koha tried to make it the case
> > that koha goes). If koha had previously gone, then {koha klama}
> > would already be true. So we need a tense on klama meaning
> > "after the trying", "thereafter", {ba *tense of superordinate bridi*}.
> I'm not sure I see the problem. What if koha had previously gone but
> somewhere else than zohe, or by some other route, why is the tense more
> important than these implicits?
 
The implicit sumti are present in the duhu/proposition if not in the
seduhu/sentence. If the tense is not in the seduhu then it's not in
the duhu. This would not be so if, as I would advocate, every or
at least most selbri had event sumti.
 
> As an aside, how do the tenses of subordinate bridi behave? Are they
> absolute, or do they refer to the main bridi?
 
I presume they're absolute. But I very very very hazily half-recall
Nick saying otherwise.
 
---
And