Re: Variables and connectives.
ma...@kli.org
>Date: Thu, 02 Sep 1999 13:13:39 PDT
>
>From: "Jorge Llambias" <jjlla...@hotmail.com>
>
>la mark cusku di'e
>
>>I figure I could avoid weird tanru and whatever by simply saying {da cmene
>>mi .e do}.
>
>I tend to avoid logical connectives and probably would
>say {mi do mintu le ka cmene}, or more explicitly
>{mi do mintu le ka makau cmene ke'a}.
(maybe {ce'u} instead of {ke'a}? Not sure of the difference. {ke'a} is
for relative clauses, right? Definitely need to download a new cmavo list;
I don't have ce'u in the list, and still have po'o in POhO).
Yes, and someone else suggested in private email {lo do cmene cu cmene mi}
which works too. I just happened to think of the predicate-logic style,
with variables, and was wondering if I had it right. No claims that it's
preferable.
>> Note that {da cmene mi .e do} is equivalent to
>>{da cmene mi .ije da cmene do}.
>
>Correct. But, {mi e do se cmene da} means something
>different! It expands as {mi se cmene da ije do se
>cmene de}. This is just like {da cmene le re prenu}
>vs. {le re prenu cu se cmene da}. The first means that
>there is something that names each of the two people,
>and the second means that for each of them there is
>something that names them.
Whoa, you blew past me there. Why does the implicit prenex work
differently here? Oh, this is along the lines of moving the variable
across negation borders, and having to apply stuff like deMorgan's law?
Since the variable occurs after the {mi e do}, to move it into the prenex
it splits somehow? I'm still pretty confused, but I'm not disagreeing.
Can someone explain this?
~mark
Jorge Llambias
>From: ma...@kli.org
>
> >{mi do mintu le ka makau cmene ke'a}.
>
>(maybe {ce'u} instead of {ke'a}? Not sure of the difference.
Yes, I meant {ce'u}. I used to use {ke'a} before {ce'u} was
invented. They serve basically the same function, one for
relative clauses and the other for properties.
> >Correct. But, {mi e do se cmene da} means something
> >different! It expands as {mi se cmene da ije do se
> >cmene de}.
>
>Whoa, you blew past me there. Why does the implicit prenex work
>differently here? Oh, this is along the lines of moving the variable
>across negation borders, and having to apply stuff like deMorgan's law?
Yes, though there are no negations in this case. The order
of universal and existential quantifiers is significant.
The conjunction is like a universal quantifier, and the
disjunction is like an existential for this purpose,
so in this example we have a universal (with .e) and an
existential quantifier (with da), and the order in which
they appear is significant.
>Since the variable occurs after the {mi e do}, to move it into the prenex
>it splits somehow? I'm still pretty confused, but I'm not disagreeing.
I'm not sure this has been explicitly stated in the book,
but I don't see how it could work any other way.
co'o mi'e xorxes
Jorge Llambias
>I figure I could avoid weird tanru and whatever by simply saying {da cmene
>mi .e do}.
I tend to avoid logical connectives and probably would
say {mi do mintu le ka cmene}, or more explicitly
{mi do mintu le ka makau cmene ke'a}.
> Note that {da cmene mi .e do} is equivalent to
>{da cmene mi .ije da cmene do}.
Correct. But, {mi e do se cmene da} means something
different! It expands as {mi se cmene da ije do se
cmene de}. This is just like {da cmene le re prenu}
vs. {le re prenu cu se cmene da}. The first means that
there is something that names each of the two people,
and the second means that for each of them there is
something that names them.
co'o mi'e xorxes
John Cowan
> Note that {da cmene mi .e do} is equivalent to
> {da cmene mi .ije da cmene do}. What I wasn't sure about, but which I
> think has already been considered and is well-known, is whether the scope
> works like I think it does. That is, in {da cmene mi .ije da cmene do} am
> I in fact asserting that we're dealing with the *same* da in both
> sentences? I think so; I thought bound variables had a fairly long scope,
> till they were rebound or at least until {ni'o} or something. If not, all
> I'm asserting is that both you and I have names.
It works the way you think, because sentence1 .ije sentence2 is in the
scope of the (implicit) prenex that binds da.
--
John Cowan co...@ccil.org
I am a member of a civilization. --David Brin
ma...@kli.org
Lojban, but it got me pondering some stuff about variables and connectives
and whatnot. Probably nothing hard to answer, but I wasn't sure.
I started off on this line of thought with the sentence "we have the same
name" (I happened to see a man walking past wearing a necklace with the
name "Mark" on it). There are a few unrelated ponderings I have on it, so
the presentation might be disjointed.
I figure I could avoid weird tanru and whatever by simply saying {da cmene
mi .e do}. OK, digression: the connective. I can't use {mi'o} since
that's equivalent to {mi joi do} which would mean that there's something
that names us as a mass. Just like {la djan. joi la djim. bevri le pipno}
means that Djan and Djim carry the piano jointly, acting as a team, massed
together, {da cmene mi joi do} would have to mean that there's a name for
the mass of I/we and you (us'ns and yous'ns). Maybe we form a famous team
or something. For all that I use {jo'u} sometimes, I'm not always positive
just how it differs from {.e}. I suppose {da cmene mi jo'u do} would mean
that there's a name that applies to each of us individually, but only when
we're somehow considered together or something weird. Maybe something like
"Half the Dynamic Duo" would be such a name. So back to {.e}, which I'm
nearly certain is the right connective here (even though I've discovered
that sumti logical connectives are actually Right less often than we
think). End digression. Note that {da cmene mi .e do} is equivalent to
{da cmene mi .ije da cmene do}. What I wasn't sure about, but which I
think has already been considered and is well-known, is whether the scope
works like I think it does. That is, in {da cmene mi .ije da cmene do} am
I in fact asserting that we're dealing with the *same* da in both
sentences? I think so; I thought bound variables had a fairly long scope,
till they were rebound or at least until {ni'o} or something. If not, all
I'm asserting is that both you and I have names.
I think this business about expanding logical connectives into conjoined
sentences is part of what makes me a little unsure about the LAhE/LUhE
bracketing for attaching relative phrases/clauses to complex sumti, which
is what prompted me to ask (and which question nobody answered. Hello?)
Even if I did get confused and thought you'd need a {ro} quantifier.
Anyone?
~mark