Re: solutions to sumti opacity

[ucleaar <ucleaar@ucl.ac.uk>]

> > On this one we're going round in circles. My contention is that there
> > is no consistent definition of "want" that allows for the x2 to be
> > an object or a bridi.
> > The easiest way to convince me otherwise is to offer me such a
> > consistent definition.
>
> Ok, you accepted that vajni (important) can take both events and objects.
> Define djica based on se vajni, where the x3 is something like
> "to satisfy some urge of x1/give pleasure to x1". Maybe the definition
> can be improved using other brivla that also accept objects and events.

Fine. Eventually, one hopes, the jbobau vlacku will give more explicit
definitions of this sort.

Given the above definition for "djica", I'll go along with you with
regard to permissible x2s.

> > > And "I need the book to read it" is slightly different from "I need to
> > > read the book".
> >
> > Right. In "I need the book to read it", the desideratum is the speaker's
> > possession of the book, while in "I need to read the book" the
> > desideratum is the speaker's reading the book.
>
> And what is the difference between "it is important to read this book"
> and "this book is important as a reading material"? Isn't there a similar
> distinction?

I don't think so. But I have no idea how the semantics of "as" works,
& we're straying into English grammar here, so I won't pursue this point
any further.

> > There isn't a *preexisting* one true meaning, but words do have one
> > true meaning. We subjectively choose which meaning should be the one
> > true one. Of course with natural languages we don't make that choice,
> > since when we decide the one true meaning of a word in our idiolect
> > our decision is severely constrained by the need to ensure that the
> > meaning corresponds as closely as possible to the meaning of the word
> > in other speakers' idiolects.
> I disagree that words have one true meaning even in one's own idiolect.
> Their meaning is blurred in a more or less wide range, and depending
> on context, one area of that range is focused on.
> I agree that for Lojban, at least when defining it, what range is
> covered by the word should be made as clear as possible, but I disagree
> that giving one sharp true meaning is possible even in principle.

The one true meaning is blurry, not sharp, and is "more or less wide".
Many categories do have fuzzy boundaries, and haywire internal structure,
as with the balan category of Dyirbal, beloved of George Lakoff.
[Bob Chassell was discussing it just the other day.]

> > > >   (i) "lu broda lihu" means either "the sentence 'broda'" or
> > > >       "the thought 'broda'", or
> > >
> > > I think it means either, but it is the words, not what they express.
> >
> > Okay. Then plain "lu broda lihu" is what I want then.
> > "Do djica lu mi cliva lihu" would mean "You want to leave".
> > [Or some lujvo in "djica"'s stead.]
>
> No. It means "you want 'mi cliva'", where 'mi cliva' is an object.
> If you don't accept objects as the x2 of djica, then it doesn't make
> much sense. If you do, then what you want is that sentence, maybe
> when editing a manuscript, you decide that you want 'mi cliva' and not
> 'mi klama'.

I think this is what I want. According to my definition of djica -
which I shall now bestow instead upon a lujvo, "sizdjica", the x2
is a thought, a concept, an idea, a mental object. Since "lo siho ..."
refers to a thought, it can be an x2. According to you, "lu...lihu"
can refer to words spoken, written or thought. Thought words are
mental objects, so "lu...lihu" can be x2 of sizdjica.

> > > >   (ii) We need an analogue of "lahe" that means "the idea of", e.g.
> > > >        "xahe" in "xahe lu broda lihu", or
> > >
> > > I don't really see the difference between "the idea of" and "that which
> > > is expressed by". Can you be more explicit?
> >
> > Depending on how cognitive your semantics is, the referent of "lo mlatu"
> > is either a cat, or a concept of a cat. Whichever you prefer, this is
> > what the referent of "lahe lu lo mlatu lihu" is. If you think the
> > referent of "lo mlatu" is a cat, then I wanted the referent
> > of "xahe lu lo mlatu lihu" to be a concept of a cat. If you think the
> > referent of "lo mlatu" is a concept of a cat, then I wanted the
> > referent of "xahe lu lo mlatu lihu" to be a concept of a concept of
> > a cat.
>
> I'm confused. lu lo mlatu li'u are the words "lo mlatu". The referent is
> not a cat nor the concept of a cat, but the word "lo" followed by the
> word "mlatu".

Yes. I can't see where you might have thought I was saying otherwise.

> {la'e lu lo mlatu li'u} has as referent that which is expressed by the
> words "lo mlatu", so a cat.

Okay. That's the less cognitive of the two views I offered.

> In {la'e lu le mlatu li'u} the {le} picks up the in-mindedness of
> whoever is being cited by the speaker, because it is within quotation
> marks.

True, but the problem here is that "lahe" refers to a cat (probably),
not to a thought. So it wouldn't do for x2 of sizdjica.

---
And