Re: To be or not to be? Coffee or tea?
Logical Language Group
zasti .aipei jonai na'e zasti .aipei .i. terdji .ei
JL> Now, suppose I want to respond "either". If I say {.a}, I'm only
JL> saying that I want at least one of them, but I'm not saying which.
JL> How do I say that I want either? I would say {du'ibo}, but that's
JL> not grammatical yet.
Well my phraseology begs the question, since I don;t think it is a logical
or a true/false question, but an emotive one. But, given a ji
question
.a says either (or both!) will do
.onai says either, not both (in my version the claim is simply that one or the
othe must be true, but not both. My Hamlet knowedge is so rusty, that I am
merely intuiting that he doesn't consider both to be possible.
.enai says the first and not the second
na.e says the second and not the first
JL> mi djica le nu do pinxe loi ckafi gi'a pinxe loi tcati
JL> Does it further expand to
JL>
JL> mi djica le nu do pinxe loi ckafi kei .a le nu do pinxe loi tcati
JL>
JL> ?
JL> The first one means that I want that you drink at least one of them,
JL> but I don't have to want that you drink one in particular. In the second
JL> one, I have to want that you drink one in particular.
I don't see why. If I drink both of them, I think that the desire is still
satisfied.
But in the abstract, your question is valid. Each of those "lenu" clauses
has its own prenex, and if there were any quantifiable variables in or implied
in either lenu clause, then it is not automatically valid that you can
export an arbitrary logical connective past the prenex to the higher level
of your second example. I just don't see any hidden quantifiable variables
in your example.
lojbab
Logical Language Group
> do djica tu'a loi ckafi ji loi tcati
>
> [In the draft grammar, djica is used without tu'a in several places.
> Is this sumti raising?]
Usually. In the connectives paper, all examples are sumti-raising and have
been fixed. In the places paper, all examples are sumti-raising and have
been changed to use "viska". In the text structure paper, 7.8 is
a deliberately vague topic-comment sentence and has been left alone.
I have added a note about hidden sumti-raising.
> Now, suppose I want to respond "either". If I say {.a}, I'm only
> saying that I want at least one of them, but I'm not saying which.
> How do I say that I want either? I would say {du'ibo}, but that's
> not grammatical yet.
I don't understand the distinction you are making here. If you wish to
avoid the possibility of getting both, use ".onai".
> And to add some confusion, consider
>
> mi djica le nu do pinxe loi ckafi .a loi tcati
>
> which expands to
>
> mi djica le nu do pinxe loi ckafi gi'a pinxe loi tcati
>
> Does it further expand to
>
> mi djica le nu do pinxe loi ckafi kei .a le nu do pinxe loi tcati
>
> ?
No, no, a thousand times no! Logical connectives can't expand out of
abstractions, which are referentially opaque:
mi jinvi le du'u loi jmive zvati gi'onai na zvati la .iupiter.
I opine the-fact-that a-mass-of living-things is-at or-else isn't-at Jupiter
is true, since the embedded sentence is a tautology, but:
mi jinvi le du'u loi jmive zvati la .iupiter.
.ijonai mi jinvi le du'u loi jmive zvati la .iupiter.
I opine the-fact-that a-mass-of living-things is-at Jupiter,
or-else I opine the-fact-that a-mass-of living-things isn't-at Jupiter.
is false, since I have no evidence one way or the other ("jinvi" requires some
sort of evidence, real or fancied, unlike "krici"). This example will be moved
to the connectives paper in Section 19.
> At first sight they seem equivalent, but...
>
> The first one means that I want that you drink at least one of them,
> but I don't have to want that you drink one in particular. In the second
> one, I have to want that you drink one in particular.
Quite right.
> I hope what I wrote makes any sense to someone.
Indeed.
--
John Cowan sharing account <loj...@access.digex.net> for now
e'osai ko sarji la lojban.
Jorge Llambias
> If you want the dcihotomy tobe/not to be, how about
>
> zasti .aipei jonai na'e zasti .aipei .i. terdji .ei
I'm not sure na'e zasti is one of the choices. He's considering
existence only, but other than that it looks fine. (To keep to
Nick's number of syllables constraint I would remove one of the
aipei:
aipei zasti jonai na'e zasti i terjdi ei
[note: terjdi, not terdji]
I guess the pei is ok, since he's talking to himself.
> JL> Now, suppose I want to respond "either". If I say {.a}, I'm only
> JL> saying that I want at least one of them, but I'm not saying which.
> JL> How do I say that I want either? I would say {du'ibo}, but that's
> JL> not grammatical yet.
>
> Well my phraseology begs the question, since I don;t think it is a logical
> or a true/false question, but an emotive one. But, given a ji
> question
There I was talking about the coffee or tea problem, not Hamlet.
> .a says either (or both!) will do
If I say
mi djica loi ckafi .a loi tcati
then I'm not saying that either will do. I'm saying that at least one
of them will do, and I'm not saying which one.
host: do djica loi ckafi ji loi tcati
guest: .a
[host brings guest a cup of tea]
guest: mi na djica loi tcati i mi djica loi ckafi
[host pours tea on guest's head]
The guest was telling the truth when responding {.a}
What should she say if she really meant "either"?
And should I use tu'a with djica?
> JL> mi djica le nu do pinxe loi ckafi gi'a pinxe loi tcati
> JL> Does it further expand to
> JL>
> JL> mi djica le nu do pinxe loi ckafi kei .a le nu do pinxe loi tcati
> JL>
> JL> ?
>
> JL> The first one means that I want that you drink at least one of them,
> JL> but I don't have to want that you drink one in particular. In the second
> JL> one, I have to want that you drink one in particular.
>
> I don't see why. If I drink both of them, I think that the desire is still
> satisfied.
It's not a matter of satisfying any desire. The truth value refers to whether
the desire exists to drink a particular one, or to drink either one.
> But in the abstract, your question is valid. Each of those "lenu" clauses
> has its own prenex, and if there were any quantifiable variables in or implied
> in either lenu clause, then it is not automatically valid that you can
> export an arbitrary logical connective past the prenex to the higher level
> of your second example. I just don't see any hidden quantifiable variables
> in your example.
Even without quantifiable variables the meanings are different. I was just
surprized that in those cases you can't expand a logical connective to two
bridis.
>
> lojbab
>
Jorge
Jorge Llambias
Hamlet's "to be or not to be" thing. The first line was:
> .i pu'o zasti .ei xu .i di'u ra'i preti
I think that that line is probably Shakespeare's universally
most recognized, and that it is essential for that recognition
to have the "be-not be" contrast in any translation. Of course,
purely meaningwise it's probably unnecessary, but there's more
than meaning involved here. I also couldn't understand what
{pu'o} was doing there, and I think {preti} is not the best of
translations for this type of "question", it is the issue that
is being pointed out, rather than the interrogative statement.
(I prefer {ai} instead of {ei} too, but that's more subtle.)
I'd say something like:
i aixu zasti i xu na zasti i terjdi ia
But the subject of this post is logical connectives. I don't think
they should be used for Hamlet's question, but I also have some trouble
in general with the "decision or".
do djica tu'a loi ckafi ji loi tcati
[In the draft grammar, djica is used without tu'a in several places.
Is this sumti raising?]
Now, suppose I want to respond "either". If I say {.a}, I'm only
saying that I want at least one of them, but I'm not saying which.
How do I say that I want either? I would say {du'ibo}, but that's
not grammatical yet.
And to add some confusion, consider
mi djica le nu do pinxe loi ckafi .a loi tcati
which expands to
mi djica le nu do pinxe loi ckafi gi'a pinxe loi tcati
Does it further expand to
mi djica le nu do pinxe loi ckafi kei .a le nu do pinxe loi tcati
?
At first sight they seem equivalent, but...
The first one means that I want that you drink at least one of them,
but I don't have to want that you drink one in particular. In the second
one, I have to want that you drink one in particular.
I hope what I wrote makes any sense to someone.
Jorge
Logical Language Group
> But in the abstract, your question is valid. Each of those "lenu" clauses
> has its own prenex, and if there were any quantifiable variables in or implied
> in either lenu clause, then it is not automatically valid that you can
> export an arbitrary logical connective past the prenex to the higher level
> of your second example. I just don't see any hidden quantifiable variables
> in your example.
Prenexes and quantified variables aren't the problem; the problem is that
abstractions, like quotations, are "referentially opaque". This is
easier to see for quotation. The truth conditions for "He said 'Live or die!'"
aren't the same as those for "He said 'Live!' or he said 'Die!'".