Re: [lojban-beginners] Re: mi kakne lo bajra
Jorge Llambías
list so as not to scare away the newbies.)
On Sat, Oct 30, 2010 at 8:59 AM, Lindar <lindar...@yahoo.com> wrote:
>
> xorxes: Sumti raising is sumti raising. There's no nitpicking or
> arguing semantics about it.
"Raising" is a term from linguistics that, like so many other terms
from linguistics, is often misapplied in Lojban. If you really want to
know what "raising" is, this is a good place to start:
http://en.wikipedia.org/wiki/Raising_(linguistics)
As the article says: "Not all languages have raising verbs; English is
one that does."
And Lojban is another language that does too. Or at least it has
"raising selbri", if you don't like calling selbri "verbs".
Examples of raising selbri in Lojban are simlu, mutce, milxe, cenba,
all the measure words, in fact almost all the selbri that have a
property place are "raising selbri", since they raise an argument from
the subordinate property ka bridi into the main bridi.
Other examples of potentially raising selbri are djuno, cilre, facki,
jimpe, and all the others with a place structure involving "fact x2
about x3". The argument x3 can be raised from the subordinate clause
in x2. This rarely happens however, because the raised argument is
inconveniently located. Nobody really says:
mi facki lo du'u sralo kei ko'a
"I found out being Australian about her."
"I found out about her being Australian."
Instead of:
mi facki lo du'u ko'a sralo
"I found out that she is Australian."
Raising is just not convenient in Lojban for these propositional
attitude selbri.
(Also, it is not clear why some of them have a raising place and
others, like for example "birti", don't. Either all should have it or
none, but Lojban place structures are so full of exceptions. But
that's just an aside.)
Yet another class of selbri that could be said to involve raising are
the ones corresponding to what the wikipedia article calls "control
verbs". These are selbri like troci, snada, fliba, zukte, kakne, and
so on, where the x1 works semantically both as the first argument of
the main verb but also as the (normally first) argument of the
subordinate bridi in the x2. So in:
mi troci lo nu bajra
mi snada lo nu bajra
mi zukte lo nu bajra
mi kakne lo nu bajra
and so on, the x1 of bajra is "raised" to the x1 position of the main selbri.
> Wanting an apple for the purpose of eating
> it is still sumti raising, because it's adding an implied concept of -
> having-. That's what sumti raising is. =/
It is, in a sense, sumti raising, but not for the reason you give.
Consider these:
mi pilno lo mapku lo nu dasni
mi nitcu lo mapku lo nu dasni
mi djica lo mapku lo nu dasni
"pilno", "nitcu" and "djica" all have basically the same place
structure. (There may be others like them, for example "sazri.)
Now, we could say that in those three examples, there is a double
sumti raising, since the x1 of dasni is raised to the x1 of the main
clause, and the x2 of dasni is raised to the x2 of the main clause.
But there is nothing wrong with any of them! sumti raising is a normal
part of the Lojban grammar. Some selbri just happen to have argument
places for raised arguments. So what? Why this witch-hunt about the x2
of djica? Why doesn't anyone ever worry about the tens or maybe
hundreds of other sumti raising places that the gismu list provides?
And you didn't say what you think about "dunda lo plise". Do you
object to that too, or do you wisely ignore the gismu list comment in
that case?
mu'o mi'e xorxes
John E Clifford
from a subordinate, temporary universes of discourse into the main one. If I
say "I want for me to ride a unicorn", say, I am not at all put off by the
objection "There are no unicorns" because the unicorn I want is buried in a pair
of worlds which pertain to two different counterfactual conditonals and so have
nothing to the universe of present discourse. If I say, on the other hand
(assuming English is something like a logical language :)), "I want a unicorn
for me to ride", I seem to be saying that there are unicorns (in the present
domain) and I want one of them to ride. The claim that there aren't any is then
false, even though the interlocutor has believed it true and has not agreed to
an expansion, as required by the rules of conversation. Indeed, his remark
might well be a reminder that the universe of the dialog does not encompass
unicorns (whatever may happen in wish-worlds and the like). Further, the new
form implies that there is a unicorn I want to ride and that, even in the
expanded domain, is false, since no one unicorn is singled out by my desire, but
rather any one will do. There are other problems, about the laws of identity
and the like that this move can give rise to. So, as a general rule, don't
raise unless you are sure the referent of what you raise is already set up to be
talked about.
Instead of:
--
You received this message because you are subscribed to the Google Groups
"lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to
lojban+un...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/lojban?hl=en.
Jorge Llambías
> Raising is always a risky business, because it appears to involve moving items
> from a subordinate, temporary universes of discourse into the main one.
So it does, but there's nothing special about raising in that. There
are plenty of other ways of doing it that don't involve raising. For
example:
la djan cu jinvi lo du'u zasti kei lo cevni
"John thinks about gods that (they) exist."
By saying that, we have introduced gods into the universe of discourse
through raising.
But if we say instead:
la djan cu jinvi lo du'u lo cevni cu zasti .i la djan cu so'e roi
tavla mi lo cevni
"John thinks gods exist. John often talks to me about gods."
No raising there, but we have also introduced gods into the universe
of discourse.
Introducing things into the universe of discourse is something we all
do all the time, whenever we speak. It's part and parcel of what
speaking is all about.
> If I
> say "I want for me to ride a unicorn", say, I am not at all put off by the
> objection "There are no unicorns" because the unicorn I want is buried in a pair
> of worlds which pertain to two different counterfactual conditonals and so have
> nothing to the universe of present discourse. If I say, on the other hand
> (assuming English is something like a logical language :)),
Being a logical language has nothing to do with it. You're talking
about ontology, not about logic. "There are no unicorns, so you can't
want one" is an ontological objection, not a logical one. And a silly
one at that, from someone who thinks that it is only possible to talk
about things that exist in the real/material world.
> "I want a unicorn
> for me to ride", I seem to be saying that there are unicorns (in the present
> domain) and I want one of them to ride. The claim that there aren't any is then
> false, even though the interlocutor has believed it true and has not agreed to
> an expansion, as required by the rules of conversation.
"Sorry, there aren't any here", or "there aren't any in this world" or
"sorry, but unicorns don't exist" is a perfectly legitimate and true
answer. "Universe of discourse" is not the same as "the material
universe in which we exist".
> Indeed, his remark
> might well be a reminder that the universe of the dialog does not encompass
> unicorns (whatever may happen in wish-worlds and the like).
The universe which the dialogue is about encompasses them as soon as
they are mentioned. That of course does not mean that the universe in
which the dialogue takes place suddenly encompasses unicorns.
Unfortunately it doesn't work like that. You can't create things into
existence just by talking about them. But "the universe of the
dialogue" is ambiguous, you know it, and yet you relish bringing it up
every time. Why?
> Further, the new
> form implies that there is a unicorn I want to ride and that, even in the
> expanded domain, is false, since no one unicorn is singled out by my desire, but
> rather any one will do.
I will not get drawn into that one this time.
> There are other problems, about the laws of identity
> and the like that this move can give rise to. So, as a general rule, don't
> raise unless you are sure the referent of what you raise is already set up to be
> talked about.
In other words, "don't ever speak"? Or just "don't ever speak in Lojban"?
How do you set up something to be talked about other than by mentioning it?
John E Clifford
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Sat, October 30, 2010 11:55:24 AM
Subject: Re: [lojban] Re: [lojban-beginners] Re: mi kakne lo bajra
On Sat, Oct 30, 2010 at 12:58 PM, John E Clifford <kali9...@yahoo.com> wrote:
> Raising is always a risky business, because it appears to involve moving items
> from a subordinate, temporary universes of discourse into the main one.
So it does, but there's nothing special about raising in that. There
are plenty of other ways of doing it that don't involve raising. For
example:
la djan cu jinvi lo du'u zasti kei lo cevni
"John thinks about gods that (they) exist."
By saying that, we have introduced gods into the universe of discourse
through raising.
But if we say instead:
la djan cu jinvi lo du'u lo cevni cu zasti .i la djan cu so'e roi
tavla mi lo cevni
"John thinks gods exist. John often talks to me about gods."
No raising there, but we have also introduced gods into the universe
of discourse.
Introducing things into the universe of discourse is something we all
do all the time, whenever we speak. It's part and parcel of what
speaking is all about.
I agree that we introduce things into the universe of discourse, but not that we
do it casually, in wishes or quotations, or the like. Such a remark may be the
occasion for such a change, but not the change in and of itself. To change the
main universe of discourse requires the collaboration of the interlocutors,
which must be consciously and overtly given. Exactly what constitutes that
shift is problematic, which is another reason why raising is risky. For
example, your second example, second sentence -- which does involve raising --
probably does bring them into the discussion. But it is still open to the other
conversant to say "But there are no gods, so we should lock him away."
> If I
> say "I want for me to ride a unicorn", say, I am not at all put off by the
> objection "There are no unicorns" because the unicorn I want is buried in a
>pair
> of worlds which pertain to two different counterfactual conditonals and so
have
> nothing to the universe of present discourse. If I say, on the other hand
> (assuming English is something like a logical language :)),
Being a logical language has nothing to do with it. You're talking
about ontology, not about logic. "There are no unicorns, so you can't
want one" is an ontological objection, not a logical one. And a silly
one at that, from someone who thinks that it is only possible to talk
about things that exist in the real/material world.
Note that i did not say what you find an objection. I said that it was risky to
move something from a subordinate universe to the main one. The reason why it
is risky is that it invalidates what appear to be normally valid arguments: a
matter of logic.
> "I want a unicorn
> for me to ride", I seem to be saying that there are unicorns (in the present
> domain) and I want one of them to ride. The claim that there aren't any is
then
> false, even though the interlocutor has believed it true and has not agreed to
> an expansion, as required by the rules of conversation.
"Sorry, there aren't any here", or "there aren't any in this world" or
"sorry, but unicorns don't exist" is a perfectly legitimate and true
answer. "Universe of discourse" is not the same as "the material
universe in which we exist".
Again, I carefully didn't say any of that, unless you mean by "here" or "in this
world" the universe of discourse, in which case, I do mean that and these
constitute a rejection of the raised form (if true).
> Indeed, his remark
> might well be a reminder that the universe of the dialog does not encompass
> unicorns (whatever may happen in wish-worlds and the like).
The universe which the dialogue is about encompasses them as soon as
they are mentioned. That of course does not mean that the universe in
which the dialogue takes place suddenly encompasses unicorns.
Unfortunately it doesn't work like that. You can't create things into
existence just by talking about them. But "the universe of the
dialogue" is ambiguous, you know it, and yet you relish bringing it up
every time. Why?
The point is simply that merely saying some words not only does not bring things
into existence (which I never claimed it did) but also does not bring them into
the domain from which the referents of terms in the conversation are drawn. If
I want a unicorn, that does not mean that there has to be a unicorn in the
domain of referents in my conversation. In fact, it means that, whatever
referent there may be for the term is buried away in several stages of world
shifting (moving to new domains of reference) which do not -- unless the
conversation turns that way -- have to affect the domain of the conversation at
all (by the way, for me, the expression "universe of discourse" is not
ambiguous). And, even if it does, the new expanded domain need not contain any
of the items that were in the domain in which the wished unicorn resided.
> Further, the new
> form implies that there is a unicorn I want to ride and that, even in the
> expanded domain, is false, since no one unicorn is singled out by my desire,
>but
> rather any one will do.
I will not get drawn into that one this time.
Too, bad. It's probably the weak point in the standard case. But, of course,
it is true under the standard rules.
> There are other problems, about the laws of identity
> and the like that this move can give rise to. So, as a general rule, don't
> raise unless you are sure the referent of what you raise is already set up to
>be
> talked about.
In other words, "don't ever speak"? Or just "don't ever speak in Lojban"?
How do you set up something to be talked about other than by mentioning it?
Well, you can mention it directly (though that is, admittedly, hard to do for
things not in the basic domain) or you can otherwise indicate where you want the
discussion to go. Not every place in every relation raises problems, but a
fairly large -- and diverse -- set do and need (especially in a logical
language) to be treated carefully.
mu'o mi'e xorxes
--
Jorge Llambías
>
> To change the
> main universe of discourse requires the collaboration of the interlocutors,
> which must be consciously and overtly given.
I don't think I have ever consciously and overtly collaborated in the
changing of the universe of discourse in a conversation. If I do it
overtly, it must be unconscious, because I don't really know what the
rules for doing it are.
The only thing I can think of is by omission, i.e. by refraining from
saying "there is no such thing as X" every time someone introduces
something X new to the conversation. But that omission wouldn't count
as overt, and it's hardly conscious, since the normal, unconscious
reaction is to accept any new things mentioned as part of the universe
of discourse and go on from there.
Perhaps in technical or philosophical discussions, when asking the
interlocutor to define some specific term they are using. That might
count as negotiating the universe of discourse. Or even perhaps in an
ordinary conversation when the other person uses a word I'm unfamiliar
with. But the reaction then would never be "there are no Xs", the
reaction would be "what is an X?".
I can't imagine a negotiation to introduce unicorns or cats or
computers or teleporters into an ordinary conversation. You just talk
about them.
> For
> example, your second example, second sentence -- which does involve raising --
How does "la djan cu so'e roi tavla mi lo cevni" involve raising?
It does not involve raising by the definition of raising given in
http://en.wikipedia.org/wiki/Raising_(linguistics)
"In linguistics, raising is a form of argument control in which an
argument that belongs semantically to a subordinate clause is realized
syntactically as a constituent of a higher clause."
There is no subordinate clause in the example for any of the three
arguments realized syntactically as constituents of the higher clause
to be raised from. So you must have in mind some other definition of
"raising".
>> Indeed, his remark
>> might well be a reminder that the universe of the dialog does not encompass
>> unicorns
[...]
> "the universe of the dialogue" is ambiguous,
[...]
> (by the way, for me, the expression "universe of discourse" is not
> ambiguous).
Nor for me. But "the universe of the dialogue" can be, because it
suggests "the universe where the dialogue takes place" instead of the
universe of discourse.
I just don't see how anyone can nonchalantly make a remark that the
universe of discourse does not encompass this or that right after
someone else has mentioned it. It may make a lot of sense to remark
that the world where the dialogue takes place does not encompass it,
but remarking that the universe of discourse does not encompass it
seems completely unintuitive and counter to all rules of conversation.
Pierre Abbat
> "Raising" is a term from linguistics that, like so many other terms
> from linguistics, is often misapplied in Lojban. If you really want to
> know what "raising" is, this is a good place to start:
> http://en.wikipedia.org/wiki/Raising_(linguistics)
What are the others, and how should Lojban be discussed in proper linguistic
terms? I'm aware of "modal", which in the Book denotes a preposition derived
from a selbri (i.e., a proper preposition; "do'e", which is not derived from
a brivla, means "fi'o co'e", so it still qualifies), and in linguistics
denotes words like "ka'e".
Pierre
--
Don't buy a French car in Holland. It may be a citroen.
Jorge Llambías
> On Saturday 30 October 2010 10:27:45 Jorge Llambías wrote:
>> "Raising" is a term from linguistics that, like so many other terms
>> from linguistics, is often misapplied in Lojban. If you really want to
>> know what "raising" is, this is a good place to start:
>> http://en.wikipedia.org/wiki/Raising_(linguistics)
>
> What are the others, and how should Lojban be discussed in proper linguistic
> terms?
Good question. I don't know, I'm not a linguist, but I have learned
not to take any traditional Lojban terminology for granted. One that
came up recently is "Zipf's law".
> I'm aware of "modal", which in the Book denotes a preposition derived
> from a selbri (i.e., a proper preposition; "do'e", which is not derived from
> a brivla, means "fi'o co'e", so it still qualifies), and in linguistics
> denotes words like "ka'e".
Yes, that's another case.
Lindar
> jimpe, and all the others with a place structure involving "fact x2
> about x3". The argument x3 can be raised from the subordinate clause
> in x2. This rarely happens however, because the raised argument is
> inconveniently located.
When x1 of a subordinate clause is elided, it's assumed to be the x1
of the main bridi. That's not what we're discussing, so forgive me if
I've applied the wrong terminology.
> Nobody really says:
>
> mi facki lo du'u sralo kei ko'a
> "I found out being Australian about her."
> "I found out about her being Australian."
> Instead of:
>
> mi facki lo du'u ko'a sralo
> "I found out that she is Australian."
>
> Raising is just not convenient in Lojban for these propositional
> attitude selbri.
>
> (Also, it is not clear why some of them have a raising place and
> others, like for example "birti", don't. Either all should have it or
> none, but Lojban place structures are so full of exceptions. But
> that's just an aside.)
> > Wanting an apple for the purpose of eating
> > it is still sumti raising, because it's adding an implied concept of -
> > having-. That's what sumti raising is. =/
>
> It is, in a sense, sumti raising, but not for the reason you give.
> Consider these:
>
> mi pilno lo mapku lo nu dasni
> mi nitcu lo mapku lo nu dasni
> mi djica lo mapku lo nu dasni
> "pilno", "nitcu" and "djica" all have basically the same place
> structure. (There may be others like them, for example "sazri.)
>
> Now, we could say that in those three examples, there is a double
> sumti raising, since the x1 of dasni is raised to the x1 of the main
> clause, and the x2 of dasni is raised to the x2 of the main clause.
> But there is nothing wrong with any of them! sumti raising is a normal
> part of the Lojban grammar. Some selbri just happen to have argument
> places for raised arguments. So what? Why this witch-hunt about the x2
> of djica? Why doesn't anyone ever worry about the tens or maybe
> hundreds of other sumti raising places that the gismu list provides?
place. What about the apple do you want/need? You've expressed reason
and the target, but not what to do with it. -THAT- is my problem here.
{pilno} is a bad example, because there's nothing else implied.
{nitcu} ... do you need possession? Do you need to throw it? The
problem is that the definition does not include -having-. So do we
assume when it's an object it's having, and something else in all
other cases? That's not what Lojban is about. That's just plain bad
practice.
> And you didn't say what you think about "dunda lo plise". Do you
> object to that too, or do you wisely ignore the gismu list comment in
> that case?
transfer of an object from one person to another. It's like borrowing
a pen. It -could- mean a transfer of ownership, but no such sense is
implied by the word in of itself.
Jorge Llambías
>>
>> mi pilno lo mapku lo nu dasni
>> mi nitcu lo mapku lo nu dasni
>> mi djica lo mapku lo nu dasni
>
> Two of these are wrong. =\
And how about:
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
Is the first one wrong?
> {nitcu} and {djica} both have (or should have) an abstracted second
> place.
"abstracted" is another one of those abuses of terminology. There is
nothing abstract about wearing a hat. A number is abstract, a property
is abstract, a set is abstract, wearing a hat is not abstract. But I
understand what you mean, you say that you can only need or want
events, not objects. But then how come you can make use of objects?
Isn't it the possession of those objects that you make use of?
> What about the apple do you want/need? You've expressed reason
> and the target, but not what to do with it. -THAT- is my problem here.
I did express what to do with the hat I want: wear it.
I can do the same for the apple:
mi djica lo plise lo nu renro fi lo stedu be do
mi nitcu lo plise lo nu renro fi lo stedu be do
mi pilno lo plise lo nu renro fi lo stedu be do
> {pilno} is a bad example, because there's nothing else implied.
How come? Don't you need to have something before you can use it? How
could you make use of it if you don't have it?
mi djica lo nu tolcau lo plise kei lo nu renro py lo stedu be do
mi nitcu lo nu tolcau lo plise kei lo nu renro py lo stedu be do
mi pilno lo nu tolcau lo plise kei lo nu renro py lo stedu be do
> {nitcu} ... do you need possession? Do you need to throw it? The
> problem is that the definition does not include -having-. So do we
> assume when it's an object it's having, and something else in all
> other cases? That's not what Lojban is about. That's just plain bad
> practice.
You don't have to assume anything. Of course if what you are going to
do with it is throw it, you will need to have it, but first you will
need for it to exist, and before that you will need for the tree to
grow, and before that you will need the Earth to exist so that the
tree can grow there, and before that you will also need the Sun to
exist, and we can go on. So yes, needing something probably means that
you need it to exist and that you need to have it. So what? You still
need it.
>> And you didn't say what you think about "dunda lo plise". Do you
>> object to that too, or do you wisely ignore the gismu list comment in
>> that case?
>
> {dunda} implies no transfer of ownership. It's a simple physical
> transfer of an object from one person to another. It's like borrowing
> a pen. It -could- mean a transfer of ownership, but no such sense is
> implied by the word in of itself.
I'm not talking about ownership. I'm talking about having it. Do you say:
mi dunda lo nu tolcau lo plise kei do
or do you say:
mi dunda lo plise do
Do I just transfer the apple to you, or do I transfer the having of
the apple to you?
If what you want is the having of the apple, should I just transfer
the apple, or should I transfer the having of the apple?
The gi'uste says that "mi dunda lo plise do" is wrong or ambiguous. Do
you agree?
John E. Clifford
Sent from my iPad
On Oct 30, 2010, at 17:48, Lindar <lindar...@yahoo.com> wrote:
Other examples of potentially raising selbri are djuno, cilre, facki,
jimpe, and all the others with a place structure involving "fact x2
about x3". The argument x3 can be raised from the subordinate clause
in x2. This rarely happens however, because the raised argument is
inconveniently located.
That's not what I'm talking about. That's a feature of the language.
When x1 of a subordinate clause is elided, it's assumed to be the x1
of the main bridi. That's not what we're discussing, so forgive me if
I've applied the wrong terminology.
The case cited is not a case where x1 in the subordinate is elided because it is the same as the x1 in the main clause, but rather of some term in the subordinate clause being moved from the subordinate clause to another place in the main clause. As noted, this is rarely done when the subordinate clause comes before the place where the raising would place the term (although shifting that place is not that uncommon).
Nobody really says:
mi facki lo du'u sralo kei ko'a
"I found out being Australian about her."
"I found out about her being Australian."
We really should, though. >_>
Why? it's awkward and it can lead to problems.
Instead of:
mi facki lo du'u ko'a sralo
"I found out that she is Australian."
Raising is just not convenient in Lojban for these propositional
attitude selbri.
(Also, it is not clear why some of them have a raising place and
others, like for example "birti", don't. Either all should have it or
none, but Lojban place structures are so full of exceptions. But
that's just an aside.)
I think that's an error. I've already semi-addressed it to the BPFK.
Which? Some having and some not having raising places (we can actually provide those places -- grammatically -- for any word; they just don't make any sense in many cases) or that there are general exceptions in Lojban place structure? The latter is simply an observable fact, the former is more easy to dispute (one may argue that there are significant differences behind the different place structures -- and that there are not). And has been.
Wanting an apple for the purpose of eating
it is still sumti raising, because it's adding an implied concept of -
having-. That's what sumti raising is. =/
It is, in a sense, sumti raising, but not for the reason you give.
Consider these:
mi pilno lo mapku lo nu dasni
mi nitcu lo mapku lo nu dasni
mi djica lo mapku lo nu dasni
Two of these are wrong. =\
Careful! In the rush to avoid one (or two) sort of error, you do not want to create a new sort of error. There are surely occasions when I want or maybe even need a particular hat (in this case) and then these three sentences present no problems, even if they do involve raising (which it is not perfectly clear they do -- as opposed to elision of repeated information). The problem comes when there is not such a particular hat and we do raise it out of the event description anyhow. There is no way to tell in Lojban which of these histories the given sentence comes from (notice this is a problem even for 'pilno' if we pull the term across tense or modality borders). So, we can find out what hat he uses by looking and what hat he neds or wants by asking or some such thing. But when we can't do that, even in principle, the raising ought to be illegitimate (and marked somehow to prevent it -- putting it into some abstraction phrase: nu, ka, and the like
will work).
"pilno", "nitcu" and "djica" all have basically the same place
structure. (There may be others like them, for example "sazri.)
Now, we could say that in those three examples, there is a double
sumti raising, since the x1 of dasni is raised to the x1 of the main
clause, and the x2 of dasni is raised to the x2 of the main clause.
But there is nothing wrong with any of them! sumti raising is a normal
part of the Lojban grammar. Some selbri just happen to have argument
places for raised arguments. So what? Why this witch-hunt about the x2
of djica? Why doesn't anyone ever worry about the tens or maybe
hundreds of other sumti raising places that the gismu list provides?
Because these are grammatical changes which do not correspond to valid argument, contra "the Logical Language" (one version, anyhow), It does seem that the racial matter is cases where human intentions are involved and only for these is there a problem. And, with varying likelihoods: "I gave an apple" has almost zero chance of being a problem (or of being a raising, for that matter), "That is a picture of a horse" has a significant but not major chance of being an improper raising, "I need a hammer" is virtually certain to be improper.
{nitcu} and {djica} both have (or should have) an abstracted second
place.
Here again, you have overshot the mark. These predicates don't *require* abstract second arguments, but they usually need them to be true, But not always (see above).
What about the apple do you want/need?You've expressed reason
and the target, but not what to do with it. -THAT- is my problem here.
{pilno} is a bad example, because there's nothing else implied.
{nitcu} ... do you need possession? Do you need to throw it? The''''
problem is that the definition does not include -having-. So do we
assume when it's an object it's having, and something else in all
other cases? That's not what Lojban is about. That's just plain bad
practice.
I see your general point and it is somewhat right: we need to insulate the term here from the domain of discourse and putting it in an event description, say, does that very nicely (though that gets ignored as well). But it is the insulation, not the additional information about why one wants or needs an object, what one is going to do with it, and so on, that is required.
And you didn't say what you think about "dunda lo plise". Do you
object to that too, or do you wisely ignore the gismu list comment in
that case?
{dunda} implies no transfer of ownership. It's a simple physical
transfer of an object from one person to another. It's like borrowing
a pen. It -could- mean a transfer of ownership, but no such sense is
implied by the word in of itself.
xorxes' point (though not how we would put it) is that, if it does involve a change of ownership, then it is likely that any raising involved (I forget what the structure of dunda' is, so I am not clear where the raising come in), it may be illegitimate (in the semantic/logical sense -- it is always grammatically allowed).
There are other ways around this problem, notably taking 'lo broda' to refer not to brodas directly but to brodaness or broda-type, but that creates problems in the ordinary, otherwise nonproblematic uses of the term, which seems a net loss. (What is the type of three girls walking down the street and how is involved in causing a car wreck?).
John E. Clifford
Sent from my iPad
On Oct 30, 2010, at 19:07, Jorge Llambías <jjlla...@gmail.com> wrote:
On Sat, Oct 30, 2010 at 7:48 PM, Lindar <lindar...@yahoo.com> wrote:
mi pilno lo mapku lo nu dasni
mi nitcu lo mapku lo nu dasni
mi djica lo mapku lo nu dasni
Two of these are wrong. =\
And how about:
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
mi pilno lo nu mi tolcau lo mapku kei lo nu mi dasni lo mapku
Is the first one wrong?
{nitcu} and {djica} both have (or should have) an abstracted second
place.
"abstracted" is another one of those abuses of terminology. There is
nothing abstract about wearing a hat. A number is abstract, a property
is abstract, a set is abstract, wearing a hat is not abstract. But I
understand what you mean, you say that you can only need or want
events, not objects. But then how come you can make use of objects?
Isn't it the possession of those objects that you make use of?
But the event of wearing a hat is abstract in Lojban terminology (and semantics, since it is a type or some such notion) and I suppose that this is what he means. He is wrong, of course in insisting that these relations require such terms in their second place. I don't get your point: how do you make use of the possession of an object (unless this is a very misleading way of saying you make use of the object).
What about the apple do you want/need? You've expressed reason
and the target, but not what to do with it. -THAT- is my problem here.
I did express what to do with the hat I want: wear it.
I can do the same for the apple:
mi djica lo plise lo nu renro fi lo stedu be do
mi nitcu lo plise lo nu renro fi lo stedu be do
mi pilno lo plise lo nu renro fi lo stedu be do
{pilno} is a bad example, because there's nothing else implied.
How come? Don't you need to have something before you can use it? How
could you make use of it if you don't have it?
mi djica lo nu tolcau lo plise kei lo nu renro py lo stedu be do
mi nitcu lo nu tolcau lo plise kei lo nu renro py lo stedu be do
mi pilno lo nu tolcau lo plise kei lo nu renro py lo stedu be do
{nitcu} ... do you need possession? Do you need to throw it? The
problem is that the definition does not include -having-. So do we
assume when it's an object it's having, and something else in all
other cases? That's not what Lojban is about. That's just plain bad
practice.
You don't have to assume anything. Of course if what you are going to
do with it is throw it, you will need to have it, but first you will
need for it to exist, and before that you will need for the tree to
grow, and before that you will need the Earth to exist so that the
tree can grow there, and before that you will also need the Sun to
exist, and we can go on. So yes, needing something probably means that
you need it to exist and that you need to have it. So what? You still
need it.
And you didn't say what you think about "dunda lo plise". Do you
object to that too, or do you wisely ignore the gismu list comment in
that case?
{dunda} implies no transfer of ownership. It's a simple physical
transfer of an object from one person to another. It's like borrowing
a pen. It -could- mean a transfer of ownership, but no such sense is
implied by the word in of itself.
I'm not talking about ownership. I'm talking about having it. Do you say:
mi dunda lo nu tolcau lo plise kei do
or do you say:
mi dunda lo plise do
Do I just transfer the apple to you, or do I transfer the having of
the apple to you?
If what you want is the having of the apple, should I just transfer
the apple, or should I transfer the having of the apple?
The gi'uste says that "mi dunda lo plise do" is wrong or ambiguous. Do
you agree.
This last part seem pretty pointless to me, since it is off on something that is not the problem, about whether we have given a purpose for giving something or whether we have given possession along with -- or instead of (possible?) -- the apple and so on. The problem, to come back to it is, is whether the referent of a term is in the domain of discourse or not, In a lot of cases involving terms that construct out of human intentions and emotions and cognitions, the answer is that typically the term we put in that place is not, in fact referring to something in the domain of discourse, and should be marked accordingly. To distinguish it from the occasional cases where it does so refer, if for nothing else. Lojban decided a long time ago to deal with this problem not by marking certain places as being peculiar with respect to some rules but by using certain term types that disallowed raising. We have fairly frequently failed to follow those plans
with various weird results, but the plan is still a good one. People who say 'mi djicu lo plise' should be prepared to answer, "Which one?" and, if they cannot in principle even do that, then their claim is false.
Jorge Llambías
>
[on "dunda"]
>
> xorxes' point (though not how we would put it) is that, if it does involve a change of ownership, then it is likely that any raising involved (I forget what the structure of dunda' is, so I am not clear where the raising come in), it may be illegitimate (in the semantic/logical sense -- it is always grammatically allowed).
That doesn't sound like my point.
My point in bringing up dunda was that the gi'uste says: "x2 may be a
specific object, a commodity (mass), an event, or a property;
pedantically, for objects/commodities, this is sumti-raising from
ownership of the object/commodity", and that fortunately nobody pays
any attention to that comment. There is no sumti-raising involved in
"mi dunda lo plise do". Especially not pedantically, i.e. when using
"raising" in its standard sense.
Jorge Llambías
>
> But the event of wearing a hat is abstract in Lojban terminology
I know, that's why I say Lojban terminology is so haywired sometimes.
> (and semantics, since it is a type or some such notion)
So when I say:
mi viska lo nu do dasni lo mapku
is there some problem? Is that a different sense of "viska" from:
mi viska lo mapku
> I don't get your point: how do you make use of the possession of an object (unless this is a very misleading way of saying you make use of the object).
It's just as misleading/non-misleading as specifying that you want to
possess it, or that you need to possess it, instead of just saying
that you want it or need it. It is for the most part unnecessarily
overprecise.
> The problem, to come back to it is, is whether the referent of a term is in the domain of discourse or not, In a lot of cases involving terms that construct out of human intentions and emotions and cognitions, the answer is that typically the term we put in that place is not, in fact referring to something in the domain of discourse, and should be marked accordingly.
So you claim, but it doesn't sound right to me. It seems to me that
saying something about something is enough to put it in the domain of
discourse.
> To distinguish it from the occasional cases where it does so refer, if for nothing else. Lojban decided a long time ago to deal with this problem not by marking certain places as being peculiar with respect to some rules but by using certain term types that disallowed raising. We have fairly frequently failed to follow those plans
> with various weird results, but the plan is still a good one.
The plan was never a coherent one, and the implementation was a total
disaster, since many people are now convinced that "mi djica ta" for
"I want that" is incorrect Lojban.
> People who say 'mi djicu lo plise' should be prepared to answer, "Which one?" and, if they cannot in principle even do that, then their claim is false.
If I tell you I want an apple, I have to tell you which apple or else
my claim is false, but if I tell you I want the having of apples, I
don't need to tell you which having of apples and my claim is still
true? That's not coherent.
Lindar
thoroughly my side of the argument, and you seem to be ignoring my
statements instead of addressing them. For the record, I have
explained that the reason it's wrong is because the meaning cannot
change based on context. When {djica} means "wants" in some contexts
and "wants to have" in others, that is bad. This is why pilno/dunda is
okay and djica/nitcu is not. We're not discussing universes of
discourse, epistemology, unicorns, magical crabs, and whether we give
an apple or give ownership of an apple. It comes down to the simple
fact that a gismu cannot change meanings based on context, and you're
implying that it can. Consistency is important. If {djica lo plise}
means "Want to have an apple." then what does {djica lo nu bajra}
mean? "Want to have a running."? You're bringing up a lot of pointless
bullshit that really doesn't have anything to do with the actual
problem here. You're constantly making comparisons to English and then
justifying your malgli because it's valid in English.
You are not listening, you are not following logic or standard
practice, and I am getting extremely frustrated.
So, I am going to, as politely as possible, bow out of this
conversation.
I'm done.
Jorge Llambías
> You're just being arbitrary at this point. I've explained quite
> thoroughly my side of the argument, and you seem to be ignoring my
> statements instead of addressing them.
I thought I did address them.
> For the record, I have
> explained that the reason it's wrong is because the meaning cannot
> change based on context.
I agree with that.
> When {djica} means "wants" in some contexts
> and "wants to have" in others, that is bad.
It always means "wants", in all contexts.
> This is why pilno/dunda is
> okay and djica/nitcu is not. We're not discussing universes of
> discourse, epistemology, unicorns, magical crabs, and whether we give
> an apple or give ownership of an apple.
When pc enters a discussion, unicorns always follow him. Don't blame
me for that.
> It comes down to the simple
> fact that a gismu cannot change meanings based on context, and you're
> implying that it can.
I don't think I am.
> Consistency is important. If {djica lo plise}
> means "Want to have an apple."
It doesn't. It means "want an apple".
>then what does {djica lo nu bajra}
> mean?
It means "want running".
> "Want to have a running."? You're bringing up a lot of pointless
> bullshit that really doesn't have anything to do with the actual
> problem here. You're constantly making comparisons to English and then
> justifying your malgli because it's valid in English.
I'm saying that "want an apple" is no different from "use an apple" or
"give an apple". They may all indirectly involve the having of the
apple, but the having is not a direct part of what you are saying.
And I would be happy to have this conversation in Lojban if you
prefer, so as not to let the English interfere. I'm not basing my view
on what English does or doesn't allow.
> You are not listening, you are not following logic or standard
> practice, and I am getting extremely frustrated.
Sorry about that.
I did misunderstand you on one point. I thought your argument went
like this: "the gi'uste says that the x2 of djica has to be an event,
therefore "mi djica lo plise" is wrong."
That's why I brought up "dunda", because the gi'uste also says that
"mi dunda lo plise" is wrong, and I don't suppose you agree with that.
But what you are actually saying is more like: "from my introspecting
into the real meaning of "djica", it is clear to me that only events
can be se djica, when we say we djica an object we are in fact kidding
ourselves, because what we really djica is to have that object". Am I
right that that is your argument?
What I am saying is that when we djica something, the having of it is
as important or as secondary as in the case of using something. Yes,
in both cases when we want/use an object, we also want/use the having
of it, but that doesn't mean that we don't want/use it itself.
Both djica and pilno allow events in x2. Do you agree?
mi pilno lo nu mi se slabu lo tcadu kei lo nu tolcri lo dargu
"I use my being familiar with the city to find my way around."
mi djica lo nu mi se slabu lo tcadu kei lo nu tolcri lo dargu
"I want to be familiar with the city to find my way around."
Do you agree that both of those are correct Lojban? Assuming you do,
why would it be different for:
mi pilno lo fartci lo nu tolcri lo dargu
"I use a compass to find my way around."
mi djica lo fartci lo nu tolcri lo dargu
"I want a compass to find my way around."
Yes, I could say instead that I use my having a compass, just as I use
my being familiar with the city, or that I want my having a compass,
just as I want my being familiar with the city, but that's not
required, I can just use/want the compass itself. And I'm not relying
on the English glosses to make this argument, just consider the Lojban
alone if you prefer. Of course you can always make it more precise by
saying that what you use/want is having the compass in your hand, or
somehow available to you. But if you don't specify that, you can still
use/want the compass to find your way around. And in many cases you
don't need to specify it.
> So, I am going to, as politely as possible, bow out of this
> conversation.
>
> I'm done.
OK. If I'm missing some part of your argument it is not because I'm
purposefully ignoring it.
John E Clifford
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Sun, October 31, 2010 7:17:05 AM
Subject: Re: [lojban] Re: mi kakne lo bajra
On Sun, Oct 31, 2010 at 12:56 AM, John E. Clifford <kali9...@yahoo.com> wrote:
>
> But the event of wearing a hat is abstract in Lojban terminology
I know, that's why I say Lojban terminology is so haywired sometimes.
**I don't see the problem here; "abstract term" means one whose selbri (God I
hate Lb terminology) is constructed using cmavo of a certain kind (or array of
kinds).
> (and semantics, since it is a type or some such notion)
So when I say:
mi viska lo nu do dasni lo mapku
is there some problem? Is that a different sense of "viska" from:
mi viska lo mapku
**No, but it is looking at a different object, in this case allowing for
delusions or or other sorts of misseeings. If you're sure your perception there
is veridical (love slipping that word in from time to time) then go ahead and
raise. You still may be wrong, of course, but that was always a risk. 'nu' is
probably not the best choice for an abstractor here.
> I don't get your point: how do you make use of the possession of an object
>(unless this is a very misleading way of saying you make use of the object).
It's just as misleading/non-misleading as specifying that you want to
possess it, or that you need to possess it, instead of just saying
that you want it or need it. It is for the most part unnecessarily
overprecise.
**These are not misleading, they are just different (like the 'viska' case
above). In the case of 'pilno', which is about as close to a zero probability
of problems of this sort as you can get, it is hard to figure out what the
sentence with an event term as object might mean (without a lot of context: "He
used his having a hammer as an excuse for going back to the workroom", say).
With 'djica' or 'netci' it is easier to see the point.
> The problem, to come back to it is, is whether the referent of a term is in
>the domain of discourse or not, In a lot of cases involving terms that
>construct out of human intentions and emotions and cognitions, the answer is
>that typically the term we put in that place is not, in fact referring to
>something in the domain of discourse, and should be marked accordingly.
So you claim, but it doesn't sound right to me. It seems to me that
saying something about something is enough to put it in the domain of
discourse.
**And so it is, which is why we carefully don't say anything about stuff that is
not there to be talked about, but rather bury them away in other worlds
altogether. Abstractors and (real) modals (including tenses) have the effect of
removing reference from the present domain to another (or, as Frege insisted, of
moving from present reference to present sense, which ultimately comes to the
same thing).
> To distinguish it from the occasional cases where it does so refer, if for
>nothing else. Lojban decided a long time ago to deal with this problem not by
>marking certain places as being peculiar with respect to some rules but by using
>certain term types that disallowed raising. We have fairly frequently failed to
>follow those plans
> with various weird results, but the plan is still a good one.
The plan was never a coherent one, and the implementation was a total
disaster, since many people are now convinced that "mi djica ta" for
"I want that" is incorrect Lojban.
**Well, it is true that the need to be careful in these places has been
overstressed, with the results you report, that doesn't mean the plan was a bad
one nor incoherent. It may be that the choice of abstractors to use was wrong
(I personally think it all comes down to propositions, since I am reasonably
sure they exist and am much less sure about any of the others). Of course, 'mi
djica ta' and even 'mi djjica lo mapku' are perfectly fine Lojban, but they have
consequences, one of which is the need to be able at least in principle to point
out the object intended. Clearly there is no problem with pointing out ta; you
just did. lo mapku may be somewhat harder but also may be possible (although a
purist might insist that you use 'le' in that case).
> People who say 'mi djicu lo plise' should be prepared to answer, "Which one?"
>and, if they cannot in principle even do that, then their claim is false.
If I tell you I want an apple, I have to tell you which apple or else
my claim is false, but if I tell you I want the having of apples, I
don't need to tell you which having of apples and my claim is still
true? That's not coherent.
**Well, yes. Because having an apple is a type not a token (well, it is a token
of having, but that is another story). Grammatically simple expressions are
often logically (semantically) complex. So "I want an apple" comes out as
something like "I have a felt lack such that both if I were to have (or whatever
predicate you want here) an apple, that lack would be filled and if that lack
were to be filled, I would have an apple" Now, if I know that there is an apple
in this world that does for both these buried quantifiers, I can pop back. If
not, the quantifiers stay buried under at least two world shifts.
Michael Turniansky
something? In any case, Lindar, when you say "then what does {djica
lo nu bajra} mean? "Want to have a running."?" I answer unequivocally
"Yes, that's PRECISELY what it means. I want to have a running (with
presumably me in the x1 place)". As far as Xorxes' contention
"because the gi'uste also says that "mi dunda lo plise" is wrong, and
I don't suppose you agree with that."
"x2 may be a specific object, a commodity (mass), an event, or a
property; pedantically, for objects/commodities, this is sumti-raising
from ownership of the object/commodity "
100% RIGHT ("it may be an object or commodity"), but what that
actually _implies_ from a lofty, ivory tower linguistic POV that
theoreticians who don't use language to communicate, but simple to
make niggling points that make ordinary people's heads hurt (see
above), and prevents languages from ever being spoken, so that it's a
good thing the cavemen didn't have these linguistic academicians, or
we'd still be grunting around the fire, (I'm sorry. Where was I?) is
a transfer of the ownership property. That's what "pedantically"
means. "Pendantically" means "ordinary people can ignore this fine
distinction, because if you don't you'll never get any plise".
(But of course, I've always 100% agreed with xorxes on the point
that there is absolutely no reason why the x2 of djica can't be an
object, just like the x2 of nitcu, and that the distinction between
THOSE words are arbitrary.)
--gejyspa
John E Clifford
error that both xorxes and I agree is wrong-headed, I will answer. It is not
that 'nitci' changes its meaning, it is that the objects of 'nitci' are of
different types and this difference has consequences. When the object is a
reference to an ordinary thing, this means that a raising has taken place
legitimately. If the object is an abstraction, then the raising is not (treated
as) legitmate. Of course, either of these moves may be wrong -- the raising,
though carried out, was illegitimate or the raising could have been carried out
but wasn't (a much less awful move). 'nitci' always means something like "x1 has
a felt lack such that both if (certain conditions specified in x2) were then the
lack would be filled and if the lack were filled, the the conditions would
obtain." The strange case of 'mi nictci lo plise' arises when we know (contrary
to the contrary-to-fact conditionals) that a particular apple in this world that
does the trick.
I'm done.
--
tijlan
> For the record, I have
> explained that the reason it's wrong is because the meaning cannot
> change based on context. When {djica} means "wants" in some contexts
> and "wants to have" in others, that is bad.
I see no significant change in the predicate's meaning. In both {mi
djica lo plise} and {mi djica lo nu ponse lo plise}, the meaning of
{dunda} remains "want something". Both {lo plise} and {lo nu ponse lo
plise} are something.
> Consistency is important. If {djica lo plise}
> means "Want to have an apple." then what does {djica lo nu bajra}
> mean?
{djica lo plise} doesn't mean "want to have an apple"; it means "want
<an apple>". "to have" is not a prescriptive element.
{djica lo nu bajra} means "want <to run>". The definition of {djica}
is identical to the other example.
Both {lo plise} and {lo nu bajra} are a valid djica2, in my opinion.
John E Clifford
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Sun, October 31, 2010 8:55:21 AM
Subject: Re: [lojban] Re: mi kakne lo bajra
On Sun, Oct 31, 2010 at 9:47 AM, Lindar <lindar...@yahoo.com> wrote:
> You're just being arbitrary at this point. I've explained quite
> thoroughly my side of the argument, and you seem to be ignoring my
> statements instead of addressing them.
I thought I did address them.
> For the record, I have
> explained that the reason it's wrong is because the meaning cannot
> change based on context.
I agree with that.
> When {djica} means "wants" in some contexts
> and "wants to have" in others, that is bad.
It always means "wants", in all contexts.
> This is why pilno/dunda is
> okay and djica/nitcu is not. We're not discussing universes of
> discourse, epistemology, unicorns, magical crabs, and whether we give
> an apple or give ownership of an apple.
When pc enters a discussion, unicorns always follow him. Don't blame
me for that.
**Sorry 'bout that, but I think I inherited them from an earlier discussion (I
know I didn't come up with 'pavyseljirna').
> It comes down to the simple
> fact that a gismu cannot change meanings based on context, and you're
> implying that it can.
I don't think I am.
**You aren't but not for the reasons you give
> Consistency is important. If {djica lo plise}
> means "Want to have an apple."
It doesn't. It means "want an apple".
** Right (but a particular one, though I know you disagree)
>then what does {djica lo nu bajra}
> mean?
It means "want running".
**Probably "I want to run", with subject elision.
> "Want to have a running."? You're bringing up a lot of pointless
> bullshit that really doesn't have anything to do with the actual
> problem here. You're constantly making comparisons to English and then
> justifying your malgli because it's valid in English.
I'm saying that "want an apple" is no different from "use an apple" or
"give an apple". They may all indirectly involve the having of the
apple, but the having is not a direct part of what you are saying.
**Well, yes and no. If you say 'mi djica lo plise', you can do everything with
that 'lo plise' that you can do with the one in 'mi pilno lo plise'. The
difference is in how you got the 'lo plise' there in the first place: if you use
an apple, there it is, you are using it. If you want an apple. where is it?
The logic of 'pilno' pretty certainly doesn't involve a pair of
contrary-to-fact conditionals, the way the logic of 'djica' does. So, for that
'lo plise' to be all alone in the argument place means that it was legitimately
raised from those remote places, i.e., it has a referent in the present domain.
Now, it is true of any apple (more or less) that it will satisfy your desire,
but, assuming that lo plise' in all this does not have a current referent, it is
not true of any apple, that, if my desire is fulfilled, have *that* apple -- I
might have another one instead.
And I would be happy to have this conversation in Lojban if you
prefer, so as not to let the English interfere. I'm not basing my view
on what English does or doesn't allow.
**English is hopelessly opaque on this, but never claims to be logical. Lojban
is not very clear on this but does have the groundwork to do it right -- and
should use it, given it does claim to be a logical language.
> You are not listening, you are not following logic or standard
> practice, and I am getting extremely frustrated.
Sorry about that.
I did misunderstand you on one point. I thought your argument went
like this: "the gi'uste says that the x2 of djica has to be an event,
therefore "mi djica lo plise" is wrong."
That's why I brought up "dunda", because the gi'uste also says that
"mi dunda lo plise" is wrong, and I don't suppose you agree with that.
**I don't see why 'dunda' has a restriction to event either (I have more trouble
imagining why it allows events).
But what you are actually saying is more like: "from my introspecting
into the real meaning of "djica", it is clear to me that only events
can be se djica, when we say we djica an object we are in fact kidding
ourselves, because what we really djica is to have that object". Am I
right that that is your argument?
**Usually, but not always. So you are both a little wrong. and a lot right.
*As I said before, this last bit is mostly off topic, since it is not the role
that events play that is crucial here.
John E Clifford
terms in logic and linguistics. And we do have "metaphor", left over from JCB,
and probably a few others. "Raising" actually seems to be one we got pretty
much right.
The problem at hand is to deal with raising as a grammatical operation, without
ruining a valid inference. If I can raise "a unicorn" from "I want to ride a
unicorn" to (for explicitness) "There is a unicorn I want to ride". The
grammar is unobjectionable, but the corresponding inference is invalid. Part of
what was originally built into the notion of "a logical language" was that
grammatical transformation should all be valid inferences, and by and large,
Loglan and Lojban have adhered to that. But not here, if the transformation is
allowed. The problem is that, in the standard theory (as vague as that may be),
mentioning a unicorn in a proposition (or whatever) does not put unicorns in the
domain of discourse (but in some other peripheral domain, reachable by various
nefarious means), while mentioning a unicorn outside such secondary contexts
does, So, the grammatical raising also moves a referent from some remote domain
to the current one, without all the nefarious means, Inferences don't expand
the universe of discourse. xorxes would solve this by having the unicorn added
as soon as it appears, pretty much regardless of context, so that the unicorn
was already in with the utterance of the first sentence. This blocks off some
reasonable responses and doesn't solve all the problems involved, so is
ultimately not acceptable. But he has held to it (in various forms and by
various dodges -- see the history of 'lo') -- for, gee, it seems like decades,
----- Original Message ----
From: Pierre Abbat <ph...@phma.optus.nu>
To: loj...@googlegroups.com
--
John E Clifford
except that 'nitci' requires a purpose and 'djica' does not, though it may have
one.
And, of course, you can say 'mi djica lo pliise' but be prepared to answer the
legitimate question, "Which one?" Being a logical language has its costs as
well as its benefits.
Jorge Llambías
> On Sun, Oct 31, 2010 at 12:56 AM, John E. Clifford <kali9...@yahoo.com> wrote:
>>
>> But the event of wearing a hat is abstract in Lojban terminology
>
> I know, that's why I say Lojban terminology is so haywired sometimes.
>
> **I don't see the problem here; "abstract term" means one whose selbri (God I
> hate Lb terminology) is constructed using cmavo of a certain kind (or array of
> kinds).
Are "lo fasnu", "lo se ckaji", "lo se djuno", "lo se viska" abstract,
by that definition? Their selbri are not constructed using cmavo of
the kind you have in mind.
>> (and semantics, since it is a type or some such notion)
>
> So when I say:
>
> mi viska lo nu do dasni lo mapku
>
> is there some problem? Is that a different sense of "viska" from:
>
> mi viska lo mapku
>
> **No, but it is looking at a different object, in this case allowing for
> delusions or or other sorts of misseeings. If you're sure your perception there
> is veridical (love slipping that word in from time to time) then go ahead and
> raise. You still may be wrong, of course, but that was always a risk. 'nu' is
> probably not the best choice for an abstractor here.
So "abstractions" can be visible. What would be the best choice of
"abstractor" here, if not "nu"?
> The plan was never a coherent one, and the implementation was a total
> disaster, since many people are now convinced that "mi djica ta" for
> "I want that" is incorrect Lojban.
>
> **Well, it is true that the need to be careful in these places has been
> overstressed, with the results you report, that doesn't mean the plan was a bad
> one nor incoherent.
I still don't see how it is coherent to require hats to always be
tokens, but allowing hat-wearings to be sometimes types and sometimes
tokens. There is nothing in the semantics of hats and hat-wearings to
warrant the distinction, nor in the syntax of "lo mapku" and "lo
mapnundasni".
> It may be that the choice of abstractors to use was wrong
> (I personally think it all comes down to propositions, since I am reasonably
> sure they exist and am much less sure about any of the others).
Would you have propositions be visible, or just desirable? If just
desirable, would they be desirable in the same way that objects can be
desirable?
Ian Johnson
Ian Johnson
John E Clifford
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Sun, October 31, 2010 3:34:00 PM
Subject: Re: [lojban] Re: mi kakne lo bajra
On Sun, Oct 31, 2010 at 11:15 AM, John E Clifford <kali9...@yahoo.com> wrote:
> On Sun, Oct 31, 2010 at 12:56 AM, John E. Clifford <kali9...@yahoo.com>
>wrote:
>>
>> But the event of wearing a hat is abstract in Lojban terminology
>
> I know, that's why I say Lojban terminology is so haywired sometimes.
>
> **I don't see the problem here; "abstract term" means one whose selbri (God I
> hate Lb terminology) is constructed using cmavo of a certain kind (or array of
> kinds).
Are "lo fasnu", "lo se ckaji", "lo se djuno", "lo se viska" abstract,
by that definition? Their selbri are not constructed using cmavo of
the kind you have in mind.
**Nice point. I suppose they are (I'm not sure about 'viska') or, better, as
you say, the use of abstraction is not very useful.
>> (and semantics, since it is a type or some such notion)
>
> So when I say:
>
> mi viska lo nu do dasni lo mapku
>
> is there some problem? Is that a different sense of "viska" from:
>
> mi viska lo mapku
>
> **No, but it is looking at a different object, in this case allowing for
> delusions or or other sorts of misseeings. If you're sure your perception
>there
> is veridical (love slipping that word in from time to time) then go ahead and
> raise. You still may be wrong, of course, but that was always a risk. 'nu' is
> probably not the best choice for an abstractor here.
So "abstractions" can be visible. What would be the best choice of
"abstractor" here, if not "nu"?
**I don't remember, something about the sense data I received and what they
would correlate with normally. Not that you can't observe an event (token), of
course. But Lojban has not been kind to token-type distinctions
> The plan was never a coherent one, and the implementation was a total
> disaster, since many people are now convinced that "mi djica ta" for
> "I want that" is incorrect Lojban.
>
> **Well, it is true that the need to be careful in these places has been
> overstressed, with the results you report, that doesn't mean the plan was a
bad
> one nor incoherent.
I still don't see how it is coherent to require hats to always be
tokens, but allowing hat-wearings to be sometimes types and sometimes
tokens. There is nothing in the semantics of hats and hat-wearings to
warrant the distinction, nor in the syntax of "lo mapku" and "lo
mapnundasni".
**Well, good. I don't like token type talk as much as you do, and "abstraction"
is misleading. So the hat or the hat-wearing you see is individual and
concrete. So, too, is the apple or the eating of an apple or the having of one
that you desire. The difference is that that concrete individual is not of this
world (or, at least, is not claimed to be usually). Now, we have chosen to mark
this by bringing in events and properties and various other things to make the
case. That choice made sense at the time and we now may be stuck with it, but
there may well be more transparent ways of doing what is needed here. By the
way, the problem is not in the semantics of 'lo mapku' or 'lo maonundasni' but
in that of 'viska' and 'djica' and the like.
> It may be that the choice of abstractors to use was wrong
> (I personally think it all comes down to propositions, since I am reasonably
> sure they exist and am much less sure about any of the others).
Would you have propositions be visible, or just desirable? If just
desirable, would they be desirable in the same way that objects can be
desirable?
**I don't think that propositions are visible and only rarely are they
desirable, but what I meant was that we could use 'ka' as a universal mark for
the sorts of things that turn up in these messy cases. That is, that what I am
experiencing bears a certain relation -- different for different predicates, but
all embedded in counterfactuals -- to a proposition. Somewhere in that
counterfactual clause (or plural) there is a 'lo broda'. When the referent of
that is in this world and can function there as in the other, then we can raise
that expression to the place occupied (at some level of the grammar) by
reference to the proposition. And no mayhem results. There are probably several
levels of caveats and quibbles disregarded here, but that is the gist.
John E Clifford
From: Ian Johnson <blindb...@gmail.com>
To: loj...@googlegroups.com
Sent: Sun, October 31, 2010 4:14:24 PM
Subject: Re: [lojban] Re: mi kakne lo bajra
Perhaps it is the static typer in me, but I would go so far as to say that it does not make sense for a place in a gismu to be able to be an abstraction or a concrete object. Objects filling a given predicate place have to fulfill certain basic, fundamental requirements to make semantic sense, and I think this is one of them. To me, even though I use "want" in this way in English, a {se djica} being a concrete object is like adding a function to a number.
John E Clifford
Subject: Re: [lojban] Re: mi kakne lo bajra
Jorge Llambías
> Perhaps it is the static typer in me, but I would go so far as to say that
> it does not make sense for a place in a gismu to be able to be an
> abstraction or a concrete object.
I generally agree with you there (but see the caveat below). The point
where we disagree though is probably that you consider events to be
"abstractions", and I don't.
Events are as concrete as concrete can be. They dwell in space-time
just like apples and unicorns do. They are not abstract like numbers
or sets or properties or propositions. The main difference between
events and objects is that events are relatively well defined in time
but fuzzy in space, whereas objects are relatively well defined in
space but fuzzy in time.
The caveat is that there are some predicates that admit of different
types in different contexts. For example:
lo vi bolci cu zmadu lo va bolci lo ka barda
li so cu zmadu li ci lo ka barda
So the x1 and x2 of "zmadu" can be abstractions, like numbers, or
objects, like balls, but it wouldn't make sense to compare the
magnitude of one type with the magnitude of the other type.
The x1 of ckaji too can be pretty much of any type, but only if the
corresponding property in x2 accepts the corresponding type.
The x3 of djuno can be pretty much of any type, it can be anything
about which something can be known.
And so on.
Ian Johnson
Jorge Llambías
> The "event is concrete" concept is interesting, but I disagree with the idea
> of lumping events in with physical objects in predicate structure. If
> anything they should be a third type unto themselves.
Why only "third"? Do you lump numbers, sets, properties and
propositions into one type? Which predicates accept numbers, sets,
properties and propositions but not physical objects?
> As for the issue with {zmadu}, the problem that it is possible to compare
> two objects as long as they are of the same type, but that comparison is
> defined for different types as well.
Right, the x3, the property, determines the kind of comparison. It is
essentially the same case as "ckaji", where the x2, the property,
determines what can go in x1. And indeed any predicate with a property
place. The type of the x1 of "mutce" will be determined by the type
accepted by the property in x2.
Ian Johnson
On Sun, Oct 31, 2010 at 6:55 PM, Ian Johnson <blindb...@gmail.com wrote:
Jorge Llambías
> And the idea that one argument of a predicate should be able to define the
> types of the others is interesting, but then you start talking about trying
> to implement a full-fledged type inference system in a spoken language if
> you want everything to really make sense. I don't know if that's reasonable
> (though it would be awesome if we pulled it off.)
That a "ka" argument determines the type of another argument is really
unavoidable.
In "ko'a broda lo ka ce'u brode", assuming that "ce'u" points to the
place where "ko'a" is (and that is determined by the meaning of
"broda") then obviously the type of ko'a will necessarily have to be
the type of the x1 of brode. Otherwise "ko'a" won't be able to have
the property "lo ka ce'u brode".
Ian Johnson
On Sun, Oct 31, 2010 at 7:22 PM, Ian Johnson <blindb...@gmail.com> wrote:
Lindar
I still am today.
((more bitchiness follows))
Seriously, stop using English to define Lojban. That's your whole
problem. You keep saying, "Yeah, it means 'want' like 'I want an
apple'" and that's -not- how it works. The English means "I want to
possess and have on my person an apple.". So if you're saying that {mi
djica lo plise} means that, then what does {mi djica lo nu bajra}
mean? I want to possess and have on my person running? It has to mean
the same thing every time. That's what the issue is exactly. If
anything the implication is existence. "I desire the existence of X."
is a more accurate English translation of the correct usage.
Addressing {nitcu}, it works the same way. However, {pilno} doesn't
mention anything or have any implication of ownership, possession, or
carrying, as you pointed out. One can use one's sense of direction
just the same as one can use a hammer. It doesn't imply any kind of
carrying, possession, or method of use, just that it is used, and for
what purpose, but not how.
Jorge Llambías
>
> Seriously, stop using English to define Lojban. That's your whole
> problem. You keep saying, "Yeah, it means 'want' like 'I want an
> apple'" and that's -not- how it works. The English means "I want to
> possess and have on my person an apple.". So if you're saying that {mi
> djica lo plise} means that, then what does {mi djica lo nu bajra}
> mean?
vi'o .i mi pilno lo glico ki'u lo nu do pilno lo glico .i .e'u ba'e do
pilno lo lojbo va'o lo nu do djica lo nu mi ciksi bau lo lojbo .i lu
ko'a djica ko'e ko'i li'u cu se smuni la'e di'e .i ko'a jinvi lo du'u
ko'a ka'e pilno ko'e ko'i .i va'o lo nu da'i ko'i fasnu kei ko'a gleki
gi'a se mansa .i na vajni fa lo du'u ko'e dacti gi'i kau fasnu
> I want to possess and have on my person running? It has to mean
> the same thing every time.
ie .i ro roi se smuni lo mintu .i .e'o ko na gasnu lo nu cfipu ma'a lo
ka pilno lo glico
> That's what the issue is exactly. If
> anything the implication is existence. "I desire the existence of X."
> is a more accurate English translation of the correct usage.
lo glico xe fanva na sai vajni .i ko ciksi bau lo lojbo
> Addressing {nitcu}, it works the same way. However, {pilno} doesn't
> mention anything or have any implication of ownership, possession, or
> carrying, as you pointed out. One can use one's sense of direction
> just the same as one can use a hammer.
je'e .i si'a da ka'e djica lo mruli .a lo nu se jinzi lo ka farna .i
da ka'e nitcu lo fartci .a lo nu catlu lo tarci
> It doesn't imply any kind of
> carrying, possession, or method of use, just that it is used, and for
> what purpose, but not how.
je'e .i si'a ge nai lo nu djica ko'a gi nai lo nu nitcu ko'a cu se
sarcu lo nu tolcau ko'a
Luke Bergen
ta'o ma se smuni lu genai broda ginai brode li'u .i mi puca jinvi lo du'u lu genai broda gi brode li'u cu drani .i mu'i ma do reroi cusku zo nai
Jorge Llambías
> ta'o ma se smuni lu genai broda ginai brode li'u
lu na broda .ije na brode li'u
> .i mi puca jinvi lo du'u lu
> genai broda gi brode li'u cu drani .i mu'i ma do reroi cusku zo nai
ji'a drani .i ku'i frica smuni .i lu na broda .i je brode li'u
John E. Clifford
Sent from my iPad
On Oct 31, 2010, at 19:29, Lindar <lindar...@yahoo.com> wrote:
I was PMSing yesterday.
I still am today.
((more bitchiness follows))
Seriously, stop using English to define Lojban. That's your whole
problem. You keep saying, "Yeah, it means 'want' like 'I want an
apple'" and that's -not- how it works. The English means "I want to
possess and have on my person an apple.". So if you're saying that {mi
djica lo plise} means that, then what does {mi djica lo nu bajra}
mean? I want to possess and have on my person running? It has to mean
the same thing every time. That's what the issue is exactly. If
anything the implication is existence. "I desire the existence of X."
is a more accurate English translation of the correct usage.
Addressing {nitcu}, it works the same way. However, {pilno} doesn't
mention anything or have any implication of ownership, possession, or
carrying, as you pointed out. One can use one's sense of direction
just the same as one can use a hammer. It doesn't imply any kind of
carrying, possession, or method of use, just that it is used, and for
what purpose, but not how.
--
John E. Clifford
Sent from my iPad
--
Luke Bergen
Jorge Llambías
> .ua mi pu jinvi lo du'u cusku lu na genai broda gi brode li'u va'o lo nu do
> djica lo nu cusku lu na broda .ije na brode li'u
lu na genai broda gi brode li'u cu mintu lu ga broda ginai brode li'u
> .i ki'e
je'e mi'e xorxes
tijlan
> The "event is concrete" concept is interesting, but I disagree with the idea
> of lumping events in with physical objects in predicate structure. If
> anything they should be a third type unto themselves.
To which of the types does the x1 of {tcima} belong?
Adam Lopresto
> vi'o .i mi pilno lo glico ki'u lo nu do pilno lo glico .i .e'u ba'e do
> pilno lo lojbo va'o lo nu do djica lo nu mi ciksi bau lo lojbo .i lu
> ko'a djica ko'e ko'i li'u cu se smuni la'e di'e .i ko'a jinvi lo du'u
> ko'a ka'e pilno ko'e ko'i .i va'o lo nu da'i ko'i fasnu kei ko'a gleki
> gi'a se mansa .i na vajni fa lo du'u ko'e dacti gi'i kau fasnu
li'u
.i lu mi goi ko'a djica lo plise goi ko'e lo nu citka ku goi ko'i li'u
zo'u drani xu pe'i dai
.i mi jinvi lo du'u mi ka'e pilno lo plise lo nu citka
.i pe'i pe'i pei lo plise ba'e na ka'e fasnu
Luke Bergen
Michael Turniansky
Jorge Llambías
> .i de'i li 2010 pi'e 10 pi'e 31 la xorxes cusku lu
>
>> ko'a djica ko'e ko'i li'u cu se smuni la'e di'e .i ko'a jinvi lo du'u
>> ko'a ka'e pilno ko'e ko'i .i va'o lo nu da'i ko'i fasnu kei ko'a gleki
>> gi'a se mansa .i na vajni fa lo du'u ko'e dacti gi'i kau fasnu
>
> li'u
>
> .i lu mi goi ko'a djica lo plise goi ko'e lo nu citka ku goi ko'i li'u
> zo'u drani xu pe'i dai
pe'i drani
> .i mi jinvi lo du'u mi ka'e pilno lo plise lo nu citka
ie je'e
> .i pe'i pe'i pei lo plise ba'e na ka'e fasnu
ie lo plise na fasnu .i lo nu citka cu ka'e fasnu
Jorge Llambías
> .ienai .i lu mi djica lo plise li'u se smuni ma
.i se smuni lo du'u do jinvi lo du'u do ka'e pilno lo plise lo se gleki be do
Luke Bergen
--
Jorge Llambías
> .uanai .i ki'u ma do jinvi tu'a zo pilno .i lo nu mi djica da kei smuni ma
> .i va'o lo nu da du lo gerku kei mi djica lo nu ponse da .ije va'o lo nu da
> du lo plise kei mi djica lo nu citka da .ijesai va'o lo nu da du lo cinse
> cinri ninmu kei mi djica lo nu gletu da .i xu do jinvi lo du'u zo pilno
> mintu zo ponse .e zo gletu .u'i
.i do tolmo'i fi lo te djica
.i do djica lo plise lo nu citka ri .i do djica lo cinse cinri ninmu
lo nu gletu ri .i je'u pei do djica lo gerku lo nu ponse ri .i .e'u ko
djica lo gerku lo nu ri pendo do
no'i ie zo pilno cu milxe lo ka nalsatci .i ku'i ko pensi lo sidju
ji'a .enai lo tutci po'o .i ko'a djica ko'e ko'i .i ko'a jinvi lo du'u
ko'e sidju ko'a ko'i
Jonathan Jones
I haven't really been following this very long thread very well, especially with the huge walls of text that keep coming up.
--
mu'o mi'e .aionys.
.i.a'o.e'e ko cmima le bende pe lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )
Ian Johnson
--
Oren
.alyn.post.
> I don't mean to hijack this conversation, but I feel like my last question
> (which I asked twice) hasn't been answered by anybody.
> It doesn't deal with the question of when/how/why there is sumti raising,
> but a rather practical note of whether we could/should explicitly state in
> some document precisely which gismu are recommended to function as things
> like events, states, objects, etc.
> Example:
> bajra - may implicitly be 'nu,' may implicitly be 'pu'u'
> My question is, is this type of information useful and valid? If so, can
> it be curated, and would anybody like to help? I started a spreadsheet
> with wiki editting privileges (no login necessary).
> co'o mi'e korbi
>
Oren,
I would find this information extremely useful. I would be able to
put it to use. I can only play a limited role in collecting it.
-Alan
--
.i ko djuno fi le do sevzi
Lindar
Where x2 of broda asks for {nu} and x1 of brode asks for {nu}, {.i
broda lo brode} is kosher, because lo brode already -is- an event. For
all other cases, an abstractor is necessary.
(barring all the other bullcrap/arguments going on right now)
xorxes, since you have all these ideas about how Lojban should work,
why don't you just make your -own- language and let it stand up to
Lojban?
Adam Lopresto
> I don't mean to hijack this conversation, but I feel like my last question
> (which I asked twice) hasn't been answered by anybody.
> It doesn't deal with the question of when/how/why there is sumti raising,
> but a rather practical note of whether we could/should explicitly state in
> some document precisely which gismu are recommended to function as things
> like events, states, objects, etc.
In what way are the parenthetical remarks not exactly what you're looking for?
djica: x1 desires/wants/wishes x2 (event/state) for purpose x3
Right there it's specifying that the x2 is an event or state. Now, we
probably need to review them and fill in anything that's missing (like
the x3 above, and nandu1). But it's not a *gismu* that has a type,
it's individual places in a selbri.
> Example:
> bajra - may implicitly be 'nu,' may implicitly be 'pu'u'
None of the places of bajra (except maybe the x4) can do that.
> My question is, is this type of information useful and valid? If so, can it
> be curated, and would anybody like to help? I started a spreadsheet with
> wiki editting privileges (no login necessary).
> https://spreadsheets.google.com/ccc?key=0AoZKIeNnTBe2dFAtNHNmWTlyeFR1TzIzaUZwMXNWeEE&hl=en&authkey=CNCr2d0B
Again, the important thing is which individual places accept what
sorts of arguments. The gismu itself just relates those places.
> co'o mi'e korbi
>
John E Clifford
he has got it right. So Lojban is his "other" language. Sorry you think this
discussion is bull-crap; it is trying to work out the ramifications of Lojvan
being a logical language, dealing with both the logical part and the language
part, and shooting for reasonable resolution where they appear to conflict.
--
Oren
So then, the concept of my spreadsheet *DOES* contain useful and valid information, but would only be complete if it were expanded to include all the 'oblique' sumti places as well?
Re:"Does it bother you that *{mi pinxe lo jubme} would also be considered semantic nonsense, because tables aren't the sort of thing that one can drink?"
If things like "agent/object" are specified in these definitions, why shouldn't all 'sensical' general classes like material states ("liquid/solid") be included as well? This is in part why I was referring to these 'classifiers' as 'tags' originally. As long as people can easily point to a construction and say "according to this sumti's implied class and that selbri's meaning, this makes no sense," I think that type of judgement should have a clear litmus test. And there's nothing stopping us. With a vocabulary of less than 1500 words, many of which fall into regular sub-classes in the thesaurus, I see no reason why we shouldn't have this resource.
So, to expand the scope here, I'm proposing that each and every sumti position in gismu definitions list explicit tags for baseline sensicality. That is, for bajra:
bajra: x1 runs on surface x2 using limbs x3 with gait x4
Now account for baseline sensicality:
x1 must-be agent...
x2 must-be material...
x3 must-be material, must-be movable-part...
x4 must-be manner...
Now let's envision that these clearly specified 'baseline sensicality tags' for sumti positions are like 'keyhole definitions' that only these explicit classes can fit. Now, each sumti position also gets any number of 'key definitions' for what it can fit into, or what sumti places it can sensically 'fill.'
x1 can-act-as agent, can-act-as moving-thing, can-act-as athlete...
x2 can-act-as general-place, can-act-as surface...
x3 can-act-as body-part...
x4 can-act-as idea...
This data/document would not only provide a richer (many-to-many) series of 'categories' for vocabulary study lists, there are a series of new applications this would allow. You could automatically gauge the degree of 'figurative language' used in a text. You could automatically generate sensical example sentences for given vocabulary (or even generate a minimal spanning sensible sentence for a set of words). You could even develop a kind of auto-complete function for a lojban-specific text-editor: as you begin to type a sumti in, a list of 'sensical' suggestions could come up in a tooltip window. If we get this data, I'd totally code that!
But I want to make sure I'm understanding the nature of this data set. Please let me know if I'm still making sense, and if I do, I'll come up with technical specs for a web interface to make this data easy to gather and manage. Maybe I'll use this as a way to learn to use github.
co'o mi'e korbi
.alyn.post.
Note that even though there are a few gismu and that universe is
closed, lujvo will need some way to be marked with these classes
as well.
-Alan
> [1]minimal spanning sensible sentence for a set of words). You could even
> develop a kind of auto-complete function for a lojban-specific
> text-editor: as you begin to type a sumti in, a list of 'sensical'
> suggestions could come up in a tooltip window. If we get this data, I'd
> totally code that!
>
> But I want to make sure I'm understanding the nature of this data set.
> Please let me know if I'm still making sense, and if I do, I'll come up
> with technical specs for a web interface to make this data easy to gather
> and manage. Maybe I'll use this as a way to learn to use github.
>
> co'o mi'e korbi
>
> On Tue, Nov 2, 2010 at 12:14, John E Clifford <[2]kali9...@yahoo.com>
> > To post to this group, send email to [5]loj...@googlegroups.com.
> > To unsubscribe from this group, send email to
> > [6]lojban+un...@googlegroups.com.
> > For more options, visit this group at
> > [7]http://groups.google.com/group/lojban?hl=en.
> >
> >
> >
> >
> > --
> > You received this message because you are subscribed to the Google
> Groups "lojban" group.
> > To post to this group, send email to [8]loj...@googlegroups.com.
> > To unsubscribe from this group, send email to
> [9]lojban+un...@googlegroups.com.
> > For more options, visit this group at
> [10]http://groups.google.com/group/lojban?hl=en.
> >
>
> --
> Oren Robinson
> (315) 569-2888
> 102 Morrison Ave
> Somerville, MA 02144
>
> --
> You received this message because you are subscribed to the Google Groups
> "lojban" group.
> To post to this group, send email to loj...@googlegroups.com.
> To unsubscribe from this group, send email to
> lojban+un...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/lojban?hl=en.
>
> References
>
> Visible links
> 1. http://en.wikipedia.org/wiki/Minimum_spanning_tree
> 2. mailto:kali9...@yahoo.com
> 3. mailto:lindar...@yahoo.com
> 4. mailto:loj...@googlegroups.com
> 5. mailto:loj...@googlegroups.com
> 6. mailto:lojban%2Bunsu...@googlegroups.com
> 7. http://groups.google.com/group/lojban?hl=en
> 8. mailto:loj...@googlegroups.com
> 9. mailto:lojban%2Bunsu...@googlegroups.com
> 10. http://groups.google.com/group/lojban?hl=en
Luke Bergen
Michael Turniansky
> My only question is, will we also be marking x1 of {gerku} as being an
> animal? lo gerku are animals, lo se skari are colors, you ve skina movies,
> citka food, pilno tools, djuno facts, and djica events.
> It is nonsensical to say {la tam cu gerku} if {tam} is a block of wood.
Oh? I think "la tam cu se kelci gerku" works just fine if tam is a
block of wood.
--gejyspa
Oren
John E Clifford
relevant category. The Battle of Waterloo is an event, but you can't tell that
from its Lojban name, which looks pretty much like "The town of Waterloo" or the
name of an honest pub, "the bucket of loo water." Some places have no
restrictions, x2 (or is it x3?) of 'tavla' for instance, since we can talk about
anything, and the same is probably true of what a picture is a picture of or
what a piece of music is about and so, Others have a fairly broad range, where
it might be easier to say what "can't" go there. And, as the discussion about
'djica' shows, sometimes the very wrong thing turns up and is perfectly sensible
-- if not always transparently.
--gejyspa
--
John E Clifford
From: Oren <get....@gmail.com>
To: loj...@googlegroups.com
Sent: Tue, November 2, 2010 3:58:36 PM
Subject: Re: [lojban] Re: mi kakne lo bajra
Lindar
John E Clifford
about 'djica' (and similar ones in other cases) and for over 100 years (at
least; I haven't checked) there has been the corresponding argument in logic and
philosophy. And the answer has generally been as your report, the objects of
desire are events. But since very few languages actually talk that way, some
need arises to explain how the talking works. And that generally comes down to
saying that the event, other than its focus, tends to be very vague, but always
there. This will not, as you say, do for a logical language so we need to carry
at least a minimal event along with our pointing to the focal object. And do in
this prevents us from making a bad logical mistake. Except in the case of
things identified in the present situation. For them, we can allow reference to
the vestigial event to drop and no ill will come of it, if we treat them like
ordinary terms, not buried ones. I suppose this is a concession to bad habits
(not just English, of course, but most natural languages) but it is, at least
one that does no harm. The 'tu'a lo plise' route. is another concession,
brought about partly by the fact that we don't generally have a very clear idea
of just what we want of the focal object: its existence or even our possession
of it in some legal or practical sense seem too little, but any more precise
exposition seems off , too. Basically, I don't think we can do any better than
that and people will keep talking in wrong-headed ways, but being understood -
somewhat gurdgingly. Their remarks aren't bad grammar, they just don't actually
say what they think they say, but that is hardly a rare experience in Lojban.
:-)
much love
--
Lindar
You're very sweet, but still wrong.
No argument won. =D
lo te djica has little to nothing to do with the problem. If I want an
apple for eating, it's -still- not specifying what is wanted about the
apple. Nothing at all! lo te djica is for specifying why it is wanted,
not in what way. {mi djica lo nu terve'u pa lo plise kei lo nu citka}
is perfectly reasonable as a use of both places. x3 is the reason
itself, not what the action is of the thing in question. If yours is
the correct version, then how do I express "I want to run in order to
be healthy."?
Jorge Llambías
>
> lo te djica has little to nothing to do with the problem.
The te djica is the place for the clause from which the se djica can be raised.
> If I want an
> apple for eating, it's -still- not specifying what is wanted about the
> apple. Nothing at all! lo te djica is for specifying why it is wanted,
> not in what way.
Probably because here isn't any special way of wanting it, you just
want an apple to eat.
> {mi djica lo nu terve'u pa lo plise kei lo nu citka}
> is perfectly reasonable as a use of both places.
Of course it is.
> x3 is the reason
> itself, not what the action is of the thing in question. If yours is
> the correct version, then how do I express "I want to run in order to
> be healthy."?
mi djica lo nu mi bajra kei lo nu mi kanro
I hope you don't think I ever said you cannot put an event in x2. Sure
you can. Both events and objects are fine there. Just as in the x2 of
pilno, or the x2 of viska, or any of the many many other places that
accept both objects and events.
Lindar
> > itself, not what the action is of the thing in question. If yours is
> > the correct version, then how do I express "I want to run in order to
> > be healthy."?
> mi djica lo nu mi bajra kei lo nu mi kanro
What happens when we omit those x1s?
{mi djica lo nu bajra kei lo nu kanro}, which would be standard
practice now.
This would end up being parallel in meaning to {mi djica fi lo nu lo
nu mi bajra kei ku kanro}
...or {mi djica fi lo nu mi kanro lo nu mi bajra}, for that matter...
So now we have to explicitly mention the x1 in order for this not to
happen?
You're saying that the x2 is raised to the x2 of the x3's clause in
djica (for {mi djica lo plise lo nu citka}. So... now instead of
desirer, desired (nu), reason (probably some kind of abstraction), we
have "desirer, desired (any), reason (abstraction, x1 is the x1 of
main bridi, x2 is x2 of main bridi" if we're making a consistent rule
here.
Is this correct?
If not, please correct me.
> I hope you don't think I ever said you cannot put an event in x2. Sure
> you can. Both events and objects are fine there. Just as in the x2 of
> pilno, or the x2 of viska, or any of the many many other places that
> accept both objects and events.
Oren
So you can't make a false statement in Lojban? Or at least not one about dogs? About the outer limit of usefulness here is to refer to the abstractors (sorry) that typically occur in each argument place, when there are somre and otherwise just leave it open.
Jorge Llambías
>
>> mi djica lo nu mi bajra kei lo nu mi kanro
>
> So now we have twenty different values of 'want'.
I'm only talking about "djica", not "want", and I don't think it has
different meanings, no. I already gave you an approximate definition
in Lojban, why don't you start from there?
> What happens when we omit those x1s?
The usual thing: you have to figure them out from context.
"mi djica lo nu bajra"
is not very different from:
"mi kakne lo nu bajra"
or:
"mi zukte lo nu bajra"
or:
"mi troci lo nu bajra"
or any of several dozen others, as far as figuring out what the x1 is.
> {mi djica lo nu bajra kei lo nu kanro}, which would be standard
> practice now.
And it is perfectly fine. Do you have any problem with it?
> This would end up being parallel in meaning to {mi djica fi lo nu lo
> nu mi bajra kei ku kanro}
Why?
> ...or {mi djica fi lo nu mi kanro lo nu mi bajra}, for that matter...
Nothing I have said implies that.
> So now we have to explicitly mention the x1 in order for this not to
> happen?
Of course not.
> You're saying that the x2 is raised to the x2 of the x3's clause in
> djica (for {mi djica lo plise lo nu citka}.
No, all I said is that one frequent case is that some argument of the
subordinate in x3 is raised into the x2 of djica. Another frequent
case is that the x1 of a subordinate in x2 is raised into the x1 of
djica. As in "mi djica lo nu bajra". It doesn't always happen. For
example "mi djica lo nu carvi" does not follow the same pattern of "mi
djica lo nu bajra". Does that bother you? There are no strict rules
for figuring out what the empty places are filled with, and you
already know that, so why do you pretend that you don't?
> So... now instead of
> desirer, desired (nu), reason (probably some kind of abstraction), we
> have "desirer, desired (any), reason (abstraction, x1 is the x1 of
> main bridi, x2 is x2 of main bridi" if we're making a consistent rule
> here.
>
> Is this correct?
No. You are calling x3 the "reason", but that is incorrect. x3 is the purpose.
Suppose you want an apple. Some *reasons* for wanting an apple may be:
(1) You are hungry. (Not a good value for x3.)
(2) You like apples. (Not a good value for x3.)
(3) You saw someone else eating one. (Not a good value for x3.)
Those are possible *reasons* for you wanting to eat an apple, they
answer "why do you want one?", but they are not purposes, they don't
answer "what do you want it for?".
Other possible reasons for wanting an apple are:
(4) You want to eat it.
(5) You want to make apple pie.
(6) You want to throw it at someone.
None of those are good vales for x3, although "to eat it", "to make
apple pie" or "to throw it at someone" alone would be.
So don't confuse the reason for wanting something (which can sometimes
be wanting something else) with the purpose, what you want it for,
what you plan to do with it in case you get it.
For example:
lo nu mi citka lo plise titnanba cu krinu lo nu mi djica lo plise lo
nu zbasu lo plise titnanba
"My wanting to eat apple pie is the reason I want an apple to make apple pie."
Luke Bergen
--
John E Clifford
Subject: Re: [lojban] Re: mi kakne lo bajra
Jorge Llambías
> But you say that {mi djica lo plise} is acceptable because it's doing some
> kind of on-the-sly sumti raising of "lo nu citka lo plise" or something,
Nothing on-the-sly about it.
> then what would be wrong with {mi djuno do}?
.i do djuno mi ma
"mi" is not a proposition. Events and objects are essentially the same
type of things, the type of things that can be seen, smelled, sensed,
touched, sensed, propositions are not that type of thing.
> If we're going to allow implicit {tu'a}'s in some places,
I never said that. I don't think it's a good idea to allow implicit
"tu'a"s at all.
> then we should allow them everywhere. {mi
> se cinri la mynapoli} should be understood to really mean {mi se cinri tu'a
> la mynapoli}.
No, it should be understood as "mi se cinri la mynapolis", perfectly
fine and no "tu'a"s involved.
> And honestly, I think I'd be ok with that. If a brivla seems like it should
> take an event in one of its slots, and instead you find lo bliku, make the
> assumption that it's really {lo nu lo bliku co'e} i.e. {tu'a lo bliku}.
> Doing this would be fine with me....
Not at all fine with me.
> IF we did it consistently.
> But to say in one breath that {mi djica lo plise} is fine, but {mi djuno do}
> is not fine seems inconsistent. And I like my lojban without
> inconsistencies.
I don't see the inconsistency. Can you write down in Lojban the
definitions of "djica" and "djuno" that you have in mind that make
those two comparable? I already gave you an (approximate) definition
for "djica". One for "djuno would be (roughly) something like:
ko'a djuno ko'e ko'i ko'o
= ko'a krici ko'e ko'i .ije ko'e jetnu ko'o
We can argue about the details, but that's the basic meaning. Neither
"ko'a krici mi" nor "mi jetnu" make sense to me. Maybe you have a
different meaning for "djuno" in mind? (And don't use the polysemous
English word "know", please.)
John E Clifford
raised to the x1 of the whole, but rather the x1 of the event phrase is omitted
because it repeats the x1 of the whole.
----- Original Message ----
From: Jorge Llambías <jjlla...@gmail.com>
To: loj...@googlegroups.com
Sent: Wed, November 3, 2010 8:08:09 AM
Subject: Re: [lojban] Re: mi kakne lo bajra
or:
or:
Why?
Of course not.
For example:
--
John E Clifford
From: Luke Bergen <lukea...@gmail.com>
To: loj...@googlegroups.com
Sent: Wed, November 3, 2010 8:24:46 AM
Subject: Re: [lojban] Re: mi kakne lo bajra
Luke Bergen
John E Clifford
From: John E Clifford <kali9...@yahoo.com>
To: loj...@googlegroups.com
Sent: Wed, November 3, 2010 9:01:36 AM
Luke Bergen
Pierre Abbat
> Well, 'djuno' typically takes propositions, not event phrases, but, since I
> think they are all really propositions anyhow, I can't come down to heavily
> on that. And, of course, I can't really object to 'mi djuno do' in the
> sense of "I know (something) about you" since I don't object to 'mi djica
> ta'. The problem with the more general 'mi djuno lo gerku', say, like 'mi
> djica lo plise' is that they so often turn out to be false when you run
> them hrough the definitions (actually, I haven't worked it out for 'djuno'
> but it seems likely to be the same sort of problem).
"I know about you" is "mi djuno fi do". "I know you" is not "mi djuno do";
it's "do slabu mi".
Pierre
--
Don't buy a French car in Holland. It may be a citroen.
Adam Lopresto
realized that we have a language that is very, very good at expressing
this sort of relationship: lojban.
With that in mind, refer you to mlismu, a random bridi generator.
Specifically, it parses a data file written in lojban that contains
relationships of the sort you're describing, and produces bridi that
match the right types.
Your "must-be" and "can-act-as" are actually flip sides of the same
coin. So instead of "x1 must-be agent", I'd say something like
.i ro bajra ka'e gasnu
(Or equivalently, the set of runners is a subset of the set of agents).
Take a look at it, particularly the fatci.txt. It's not thorough at
all, and there are some things I put in there that are known to be
false, but produced humorous results (since the whole thing started as
a joke). Hopefully, I've marked those with {je'u nai} or {zo'o}, but
some may have escaped.
http://students.cec.wustl.edu/~adam/lojban/mlismu/
Oren
Ian Johnson
mu'o mi'e latros.
Adam Lopresto
.i pe'i ro nanmu cu ba'e ka'e bruna
I may have overused implicit {ka'e}, in the name of getting funny results.
Probably all of this needs to go onto a wiki page somewhere.
Luke Bergen
Jorge Llambías
> .i je'u ro bruna cu nanmu
.i ku'i su'o gerku cu bruna gi'e nai nanmu
Luke Bergen
.iru'a ro bruna cu nakni
Jorge Llambías
> .iru'a ro bruna cu nakni
ro bruna cu nakni gi'e tunba .i je ro da poi nakni gi'e tunba cu bruna