[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lei xau-dja-sei
And:
> I don't want to flog a dead horse here, but here goes anyway...
Who knows, we might even get the horse galloping if we keep at it.
=
> > lei jmabolkei po'u lo selnatmrkimri ku joi lo skoto ku
> > joi lo selnatmrxeire ku joi lo glico
> > The footballers who are at least one Welsh + at least
> > one Scottish + at least one Irish + at least one English.
> > That works. But you probably want the individual form:
> I want a form that is independent of the gadri. I.e. a form that
> defines a class of people some of whom are Welsh, some of whom
> are Irish, some of whom are Scottish, and some of whom are English
I'm not sure I follow you. Do you want an F() that is true of all those
and only those people? There are many ways to get that, for example,
using part of the above:
me lu'a lo selnatmrkimri ku joi lo skoto ku joi =
lo selnatmrxeire ku joi lo glico
... is a member of the mass made of at least one Welsh =
+ at least one Scottish + at least one Irish + at least =
one English.
But I thought you wanted something different: You want to assert of
several people that they are all the people of that class. Making
that claim is what is difficult, not defining the class. By referring
to the people as a group with lei, it is easy to identify that group =
with the whole class. But if you use le, how can you make sure that =
you are exhausting the class? You are only saying that each individual =
belongs to the class, but not that the class is exhausted by them.
> (not "glico", incidentally - that appears to denote anglophone
> culture rather than the culture of England).
I gave up on culture words. What do you suggest for England's English?
> > le jmabolkei po'u lo selnatmrkimri a lo skoto
> > a lo selnatmrxeire a lo glico
> > Each of the footballers who is a Welsh or a Scottish
> > or an Irish or an English.
> > I think that one works, too.
> No - it would allow the footballers to all only be Welsh.
I'm not sure. It depends on how "po'u" works, which is not quite clear.
Let's consider a simpler case:
ro da po'u lo mlatu
Every x which is a cat.
That seems to me to be equivalent to {ro da poi mlatu}. I.e., it covers
all cats. Now:
ro da po'u lo mlatu a lo gerku
Every x which is either a cat or a dog.
That would cover all cats and all dogs. What about with le:
ro le danlu po'u lo mlatu a lo gerku
Each of the animals that is either a cat or a dog.
That will take all the cats and all the dogs from "the animals"
that you have in mind. But you are right, if the animals in question
don't include any dogs then that leaves you only with cats. =
I now realize that we wanted "no'u" rather than "po'u" for our
example:
le jmabolkei no'u lo selnatmrkimri a lo skoto
a lo selnatmrxeire a lo glico
Each of the footballers, who is Welsh or Scottish
or Irish or English.
And you are right, if the footballers you have in mind are all =
Welsh, the comment remains true. This is unavoidable, because in
referring to them with le, the comment applies to each of them =
individually, while the claim you want to make applies to the =
group as a whole. I don't understand how you can disregard the =
gadri choice, it seems central to the issue in this case.
> Cor! No way of doing sumti tail connection? Whyever not, I
> wonder. I've been doing it for ages. It seems like a bit of
> a gaping hole... Is there a rationale for it? =
The only true logical connection in Lojban is bridi logical
connection. Sumti and bridi tail connection are just special =
cases of that, tanru connection is not really strictly logical,
and the rest (connection of NUs and similar monstrocities) are
fairly irrelevant and mercifully little used. A case for sumti
tail connection could be made, but I think it would require =
some additional grammatical structure. I agree that it is a
natural extension, but I'm not sure I agree that its lack is =
a hole, though.
> What can one
> say instead, that is literally equivalent? Are "le da voi",
> "lei da voi" possible?
No, but "le ro da voi" and "lei ro da voi" are possible. =
Or "su'o" instead of "ro", if you prefer. "le da" doesn't =
work because it forms the structure "le da broda" =3D =
"le broda pe da".
> > le prenu cu fraso gi'a dotco
> > ije su'o le prenu cu fraso
> > ije su'o le prenu cu dotco
> > But I don't know how to condense that into a simple phrase.
> Fair enough, because you can't condense it in predicate logic,
> as far as I can tell - it has to be spelled out as you've done.
> But English does allow the condensed reading, so I thought
> maybe Lojban might have found a way.
You can condense it in Lojban easily using lei. What seems =
to be an impossibility is to make at the same time a claim
about each of the players individually and about the group =
as a whole.
=
> In a way it's a shame it
> didn't, because I bet that everyone is going to feel that
> it is too long-winded to use Lojban as a logical language.
I'm not sure that we've run into any limitation here. You may
have trouble translating directly something like:
Each of the players -- who are Welsh, Scottish, English =
and Irish -- scored a goal. =
But that is because we are giving together a property of the =
group and a property of each individual. I guess we could even
manage that:
ro lu'a lei jmabolkei no'u lo selnatmrkimri kujoi lo skoto
kujoi lo selnatmrxeire kujoi lo glico cu snada lo nunse'a
Each of [the players -- who are Welsh, Scottish, Irish =
and English --] scored a goal. =
Voil=E0!
> Indeed, Lojbab has thrown the weight of his prestigious usage
> not behind the baseline but against it and the idea that
> every utterance should be logically unambiguous. =
I am convinced that Lojban usage will be plagued with =
illogicalities. My usage certainly is, when I'm not paying
a lot of attention. But in analysing the ideal language,
we can at least try to find out what the rules should be. =
Whether we should then strive to follow them or not in =
actual usage is another question.
Jorge