WikiDiscuss

WikiDiscuss


On Sat, 25 Dec 2004 12:38:17 -0800, webmaster@lojban.org
<webmaster@lojban.org> wrote:
> Something that needs to be noted in general: we, the BPFK, made a consensus decision that we do not make rulings on ontological or metaphysical issues; that is, we will not tell you whether phrase X has meaning or validity.
...
> * lo with an outer quantifier, which is exactly the same thing as just sticking a number before an item (i.e. "mu lo bakni" == "mu bakni" == "five cows), works pretty much as before: "five things that really are cows".

Was the "really are" intended?

>From what I understand, "lo" is no longer makes a statement that a {lo
broda} "really is" a broda (or "really does" broda)... so wouldn't {mu
lo bakni} be closer to "five things that are cows", without the
"really"?

> * A fractional outer quantifier selects a portion of the group. So "pa pi re loi xa bakni cu bevri lo pipno" means that one half of some group of six cows (i.e. 3 cows) carried the piano.

Does {pa pi re} mean "one half", then? I read it as "1.2", which — as
a fraction — for me means "six-fifths". So {pa pi re loi xa bakni
....} is a bridi talking about 7.2 out of 6 cows which do something or
other. Misunderstanding on my part, or mistake in the example
sentence? (Perhaps {fi'u} was meant instead of {pi}?)

mu'o mi'e .filip.
--

Philip Newton <philip.newton@gmail.com>



posts: 2388



**************
>
> ! About xorlo
>
> "xorlo" is our (the baupla fuzykamni's) pet
> name for the gadri (articles, like "the" and
> "a" in English) BPFK proposal written mostly by
> xorxes. It has the advantage of being a
> gismu form. The gadri
> proposal
(which has now been accepted by a
> vote of 11 to 0 by the BPFK, although it is
> technically subject to future changes until we
> declare ourselves done) is quite complete, but
> rather full of technical jargon.

I won't rehearse all the reasons that this
acceptance was a Bad Idea, but we are now seeing
one part of the problem: since the proposal is
internally contradictory, everything follows from
it and thus we get "explanations" which push even
beyond the already wrongheaded changes clearly
made in the original proposal.

> Also, it's the biggest change we (the BPFK)
> have made to the language, and, God willin' and
> the creek don't rise, the biggest one we'll
> ever make. By far.
>
> Hence, a tutorial seems prudent.
>
> Something that needs to be noted in general:
> we, the BPFK, made a consensus decision that we
> do not make rulings on ontological or
> metaphysical issues; that is, we will not tell
> you whether phrase X has meaning or validity.
> That is discussion and speaker specific, and
> not our job. In some discussions, saying "mi
> kalte pa lo pavyseljirna" (which litterally
> means "there exists one thing that is a
> unicorn, and I am hunting it")

That's not what it means literally at all. That
pa lo pavyseljirna exists is at most a pragmatic
implication, which, given the usual meaning of
{kalte}, does not go through. Because Lojban
does not mark its intensional contexts well --
and every effort to get it to do so has been
rejected by the same people who keep mucking up
with them — this expression does allow the
inference to something like {pa da poi
pavyseljirna se kalte mi} which does mean the
claimed literal meaning but which can be false
even when the original (as usually understood) is
true . The language — putatively logical in
some way — thus justifies and invalid inference,
not a very logical thing to do

>is perfectly
> reasonable, in others it's a reason to put
> someone in a mental hospital. In a similar
> vein, "lo" is now completely generic. This
> means that there are going to be disagreements
> about how broad it can be.

Hey, there are arguments about what, if anyhting,
that means at all. If we are going to make major
changes, we ought at least be clear what the Hell
they are. As it stands, the calim that something
is generic has been used to justify a number of
incompatible and impossible claims. Perhaps,
rather this primer should start with a definition
or at least an explanation of what "generic"
means — and "completely gneric" even more so.
As it stands, this is a loophole which might be
used to drive any truck one wanted through (and
has been).


For example, I
> think that "bear goo" is perfectly validly "lo
> cribe". Arnt does not. That's OK, albeit
> somewhat obnoxious should I ever need to
> talk about "bear goo". Of course, I can just
> use "lo pesxu be lo cribe".

{loi spisa be lo cribe} makes more sense for what
"bear goo" usually means in this context.


> !! General Notes
>
> If you choose to read the proposal itself,
> there are a couple of things you should know.
> If you just want the high-level overview, and
> have no intention of reading the proposal, skip
> this section.
>
> * "distributively" means "not as a group", and
> is a term we owe largely to
> McKay,
> whom we should give money to or something.
> Basically, "three men carried the piano" when
> handled distributively means that they each
> carried it. lo, le, and la are all
> distributive. The outer quantifier of loi,
> lei and lai is distributive over groups of
> number indicated by the inner quantifier.

Do what? Can you give an example of what this
might possibly mean — or better what it really
is intended to mean. I have a problem also (ib
addition to the muddled expression here) with the
notion of a quantifier being distributive, though
I think it may come clear when the rest of this
passage is laid out (and I have read McKay and
discussed with him about some things in his
system even).

> * "non-distributively" means "as a group".
> "Three men carried the piano" when handled
> distributively means that they all did it.

That is they all did it *together* no one — or
even two — of them did it on his (their) own.

> * There are no default quantifiers. At all.
> For example, the default outer quantifier of
> "lo" used to be "su'o", which means "at least
> one", but that is no longer the case. "lo
> cribe" could be one, or a billion, or none
> (although expect listener hostility!!), or the
> idea of bear-ness (as in "bears like honey"),
> or bear goo (as in after a car accident
> involving a really, really big truck).

This is an incredibly huge move away from clarity

and precision toward hopeless muck. If I cannot
tell whether you are talking about a thing or a
concept or a mass (in the real — non-Lojban --
sense), how am I to judge the truth of what you
say, since I don't know what you may well claim
to have said (a claim I can't check). Context
only helps so far and assumes that some things in
the context are settled; this leaves it all up in
the air. It is also not even hinted at (except
in saying {lo} is generic) in the original
proposal — either version.

> * A side effect of the above is that in xorlo,
> if you mean "one bear", consider actually
> saying "pa cribe". It's ever so much more
> specific.

Nu? Of course specifying the exact number is
more specific than not specifying. What does
that have to do with the case? Does {lo cribe}
mean "one bear" or "one concept of a bear" or
"one puddle of bear goo?" Hopefully only the
first.

>We (those of us that have actually
> been using xorlo for the last few months; there
> are at least half a dozen active users on
> #lojban now) have found that context is almost
> always sufficient, however.
>
> !! lo
>
> lo is where the biggest changes occured. In
> fact, it's fair to say that everything but the
> changes to lo (and to default quantification)
> were mere clarifications. Here's how lo works
> now:
>
> * lo is the default gadri; if in doubt, use lo
> * lo with no outer or inner quantifier is
> absolutely generic; "lo broda" means
> "something(s) or other to do with broda", and
> that's about it. Thankfully, context is plenty
> 99% of the time.

An optimistic estimate.

> Expect to see a lot more lo!
> * In particular, you almost always want "lo nu"
> rather than "le nu". "lo nu" is "some event of
> ...", "le nu" is "some particular event of ...
> that I have in mind".
> * lo with an outer quantifier, which is exactly
> the same thing as just sticking a number before
> an item (i.e. "mu lo bakni" == "mu bakni" ==
> "five cows), works pretty much as before: "five
> things that really are cows"

Whoa, Nelly. If {mu lo bakni} is just the same
as {mu bakni} then it does not mean anything
like that but rather "five things that are
somehow related to cows", possibly a cow, a cow
pie, a side of beef, a quart of milk, and
cowness. You can't have it both ways, unless you
want to say that {mu bakni} really is different
from {mu lo bakni} (which might make sense for
other reasons as well).

> * lo's inner quantifier indicates the number if
> things we're talking about, but in a slightly
> different fashion. "mu lo bakni cu bevri lo
> pipno" means "There were 5 cows; each of them
> carried a piano individually". "lo mu bakni cu
> bevri lo pipno" is ambiguous as to whether they
> did it individually or as a group. To be clear
> about group-ness, use loi and friends.

While having a vague (not really ambiguous: it
doesn't mean both, just fails to mean either)
expression is useful, it isalso useful to have a
very clear distributive expression, not
necessarily involving quantifiers — a job {lo}
used to do. Note by the way that no diddling
with the form of sumti will ever completely cover
the distributive-nondistributive distinction.

> * The above is actually a substantial change;
> "lo mu bakni cu bevri lo pipno" used to mean
> "All the cows in the universe, of which there
> are 5, carry the piano". That sucked. It is
> still possible to say the above in xorlo, but I
> don't remember the easy way off the top of my
> head; someone please replace this with an
> example of that.

Yes, this is not a wrongheaded idea!

> !! le
>
> le is basically unchanged. Because it now
> carries more baggage then lo, rather than
> less as before, it is no longer the default
> choice for the discerning Lojbanist. In my
> post-xorlo writings, lo outnumbers le by about
> three to one (at a guess). I only use le when
> I'm talking about a specific item.

That was always the rule, wasn't it? I think
there was some dispute about just where
specificity kicked in but that is about all.

> * le is used for particular things you have in
> mind.
> * Because you have them in mind in your own
> mind, it implies that your definition of
> whatever (i.e. bakni in "le bakni") may not
> agree with everyone elses (hence the famous "le
> nanmu cu ninmu" for a crossdresser example).

So, {lo} is always appropriate whenever {le}
because by using {le bakni} you have established
that the thing has something to do with cows.
This is another change-- or rather another odd
consequence of the earlier mentioned changes.

> * If you wish to use le and not have the
> implication that you might be messing with your
> user's head, you can use "le je'u <whatever>",
> but in practice we all assume that you're not
> being a jerk and that the selbri after le
> actually matches reality as you understand it.
> * The outer and inner quantifiers of le act
> exactly like xorlo for most purposes.
>
> !! la
>
> la is unchanged save for clarification.
>
> * la has no inner quantifier; a number after la
> is considered part of the name.
> * la's outer quantifier is just like lo's.
>
> !! The lVi Series
>
> This is loi, lei, and lai. They act exactly
> like lo, le and la, respectively, except that:
>
> * They make things into groups (aka masses, aka
> non-distributive groups) for purposes of the
> rest of the bridi. For example, "loi mu bakni
> cu bevri lo pipno" definately means that
> all the cows carried the piano together, as a
> group.

Since in the real world a nondistributive group
is a very narrowly specified mathematical object
and a mass is a fairly clearly defined linguistic
concept, it turns out to be a good idea to call
what we have here bunches (or something else of
that sort).

> * The inner quantifier (which lai does not
> have) indicates the size of the group.
> * You almost always want to use an inner
> quantifier with loi and lei, not an outer one.
> This may take a bit of getting used to.

Why?

> * The outer quantifier gives a number of
> groups. These are not then grouped together!
> This means that "re loi mu bakni cu bevri lo
> pipno" means that there are two groups of five
> cows, and that each group of five cows carried
> the piano.

A major change and one for which an adequate
(indeed any) justification has never been
presented (except that one person used it that
way a lot, possibly originally by mistake).

> * That example should give an idea of the power
> of xorlo; some very specific things can be
> said in xorlo very easily.

Frinstance? The description here sugests that
saying anything specific is going to be very
hard, much harder than saying some of the "new
easy expressions" in the old system.

> * Note that it is not necessarily the case that
> those two groups of five cows are completely
> distinct. They could share some members in
> common. Using this fact without making it
> clear to your listener you are doing so,
> however, is very poor form.
> * A fractional outer quantifier selects a
> portion of the group. So "pa pi re loi xa
> bakni cu bevri lo pipno" means that one half of
> some group of six cows (i.e. 3 cows) carried
> the
message truncated



posts: 2388


wrote:

> On Sat, 25 Dec 2004 12:38:17 -0800,
> webmaster@lojban.org
> <webmaster@lojban.org> wrote:
> > Something that needs to be noted in general:
> we, the BPFK, made a consensus decision that we
> do not make rulings on ontological or
> metaphysical issues; that is, we will not tell
> you whether phrase X has meaning or validity.
> ...
> > * lo with an outer quantifier, which is
> exactly the same thing as just sticking a
> number before an item (i.e. "mu lo bakni" ==
> "mu bakni" == "five cows), works pretty much as
> before: "five things that really are cows".
>
> Was the "really are" intended?

In one part of the proposal, "really are" is all
that separates {lo} from {le}, in another this is
not the case, though not to the extent it is in
the unfolding of the proposal (after it has been
acdcepted).

> From what I understand, "lo" is no longer makes
> a statement that a {lo
> broda} "really is" a broda (or "really does"
> broda)... so wouldn't {mu
> lo bakni} be closer to "five things that are
> cows", without the
> "really"?

How can something be a cow and not really be a
cow. The new expplanation of the proposal is
that it is five things that are related to cows,
not necessarily cows at all, let alone really
ones.
But partt of the proposal does als say that they
are really cows.




posts: 14214

On Sun, Dec 26, 2004 at 04:03:32PM +0100, Philip Newton wrote:
> On Sat, 25 Dec 2004 12:38:17 -0800, webmaster@lojban.org
> <webmaster@lojban.org> wrote:
> > Something that needs to be noted in general: we, the BPFK, made
> > a consensus decision that we do not make rulings on
> > ontological or metaphysical issues; that is, we will not tell
> > you whether phrase X has meaning or validity.
> ...
> > * lo with an outer quantifier, which is exactly the same thing
> > as just sticking a number before an item (i.e. "mu lo bakni" ==
> > "mu bakni" == "five cows), works pretty much as before: "five
> > things that really are cows".
>
> Was the "really are" intended?
>
> >From what I understand, "lo" is no longer makes a statement that
> >a {lo
> broda} "really is" a broda (or "really does" broda)... so wouldn't
> {mu lo bakni} be closer to "five things that are cows", without
> the "really"?

You're conflating {mu lo bakni} with {lo bakni}. The former has an
outer quantifier, the latter does not. The meanings are not the
same. It is assumed that you wouldn't talk about a number of cows
unless you actually wished to talk about cows.

> > * A fractional outer quantifier selects a portion of the
> > group. So "pa pi re loi xa bakni cu bevri lo pipno" means that
> > one half of some group of six cows (i.e. 3 cows) carried the
> > piano.
>
> Does {pa pi re} mean "one half", then?

No, that was a mistake. I meant "pi mu".

-Robin


On Monday 27 December 2004 00:47, Robin Lee Powell wrote:
> On Sun, Dec 26, 2004 at 04:03:32PM +0100, Philip Newton wrote:
> > Does {pa pi re} mean "one half", then?
>
> No, that was a mistake. I meant "pi mu".

Does it matter whether one says {pimu} or {fi'ure}, or {pira'epavorebimuze} or
{fi'uze}?

phma
--
Maintenant, j'ai besoin d'une loupe pour trouver mes lunettes!
-Les Perles de la médecine


posts: 14214

On Mon, Dec 27, 2004 at 12:57:10AM -0500, Pierre Abbat wrote:
> On Monday 27 December 2004 00:47, Robin Lee Powell wrote:
> > On Sun, Dec 26, 2004 at 04:03:32PM +0100, Philip Newton wrote:
> > > Does {pa pi re} mean "one half", then?
> >
> > No, that was a mistake. I meant "pi mu".
>
> Does it matter whether one says {pimu} or {fi'ure}, or
> {pira'epavorebimuze} or {fi'uze}?

xorxes believes that we have only handled the first (and possibly
the third, but I have *no* idea what it means). He believes, IIRC,
that we should handle fi'u as an outer quantifier in the section for
fi'u, and I care not at all so I've accepted that.

-Robin


On Monday 27 December 2004 01:03, Robin Lee Powell wrote:
> On Mon, Dec 27, 2004 at 12:57:10AM -0500, Pierre Abbat wrote:
> > Does it matter whether one says {pimu} or {fi'ure}, or
> > {pira'epavorebimuze} or {fi'uze}?
>
> xorxes believes that we have only handled the first (and possibly
> the third, but I have *no* idea what it means). He believes, IIRC,
> that we should handle fi'u as an outer quantifier in the section for
> fi'u, and I care not at all so I've accepted that.

li pira'epavorebimuze du li fi'uze. What about {cifi'ure}? I think, if {fi'ure
loi xa bakni} means half of a group of six cows, and {re loi xa bakni} means
two groups of six cows, {cifi'ure loi xa bakni} should mean one group of six
cows and half of another group.

phma
--
Without glasses, I can't even distinguish smells...
-Les Perles de la médecine


posts: 14214

On Mon, Dec 27, 2004 at 01:38:54AM -0500, Pierre Abbat wrote:
> On Monday 27 December 2004 01:03, Robin Lee Powell wrote:
> > On Mon, Dec 27, 2004 at 12:57:10AM -0500, Pierre Abbat wrote:
> > > Does it matter whether one says {pimu} or {fi'ure}, or
> > > {pira'epavorebimuze} or {fi'uze}?
> >
> > xorxes believes that we have only handled the first (and
> > possibly the third, but I have *no* idea what it means). He
> > believes, IIRC, that we should handle fi'u as an outer
> > quantifier in the section for fi'u, and I care not at all so
> > I've accepted that.
>
> li pira'epavorebimuze du li fi'uze.

No, actually. One is a number, the other is a fraction that
represents that number, *sort* of. "li fi'u ze" is "one over seven";
the fraction you gave is "li pa fe'i ze". To what extent these
things are identical is unclear. Certainly they are different
words, at the least.

> What about {cifi'ure}? I think, if {fi'ure loi xa bakni} means
> half of a group of six cows,

That's what xorxes wants to argue. I don't know what he wants "fi'u
re loi bakni " to mean (although I expect "one out of every two",
which is not the same as "half", exactly).

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"
Proud Supporter of the Singularity Institute - http://singinst.org/


On Monday 27 December 2004 01:54, Robin Lee Powell wrote:
> On Mon, Dec 27, 2004 at 01:38:54AM -0500, Pierre Abbat wrote:
> > li pira'epavorebimuze du li fi'uze.
>
> No, actually. One is a number, the other is a fraction that
> represents that number, *sort* of. "li fi'u ze" is "one over seven";
> the fraction you gave is "li pa fe'i ze". To what extent these
> things are identical is unclear. Certainly they are different
> words, at the least.

They are both numbers ({fi'u} and {ra'e} are both in PA), and they are the
same number. If I wanted to say that the *symbols* used to represent the same
number are different, I'd say {me'o pira'epavorebimuze na du me'o fi'uze}.

phma
--
My monthly periods happen once per year.
-Les Perles de la médecine


posts: 1912



> Does it matter whether one says {pimu} or {fi'ure}, or {pira'epavorebimuze}
> or
> {fi'uze}?

I propose to use {PA1 fi'u PA2 sumti} to mean
"PA1 out of every PA2 referents of sumti", whereas
{pi PA sumti} is reserved for "A piPA fraction of one
of the referents of sumti".

That two number expressions have the same li value does not
mean that they can't represent different quantifiers.
A quantifier is not a number, it is a bridi operator.

mu'o mi'e xorxes




__
Do you Yahoo!?
Yahoo! Mail - 250MB free storage. Do more. Manage less.
http://info.mail.yahoo.com/mail_250


Pierre Abbat scripsit:

> Does it matter whether one says {pimu} or {fi'ure} ...?

In a context of exact numbers, no. Where measurements are in question, .5
suggests an accuracy of one decimal place, whereas 1/2 suggests perfect
(unattainable) accuracy.

--
John Cowan jcowan@reutershealth.com
At times of peril or dubitation, http://www.ccil.org/~cowan
Perform swift circular ambulation, http://www.reutershealth.com
With loud and high-pitched ululation.


posts: 2388


<rlpowell@digitalkingdom.org> wrote:

> On Sun, Dec 26, 2004 at 04:03:32PM +0100,
> Philip Newton wrote:
> > On Sat, 25 Dec 2004 12:38:17 -0800,
> webmaster@lojban.org
> > <webmaster@lojban.org> wrote:
> > > Something that needs to be noted in
> general: we, the BPFK, made
> > > a consensus decision that we do not
> make rulings on
> > > ontological or metaphysical issues; that
> is, we will not tell
> > > you whether phrase X has meaning or
> validity.
> > ...
> > > * lo with an outer quantifier, which is
> exactly the same thing
> > > as just sticking a number before an item
> (i.e. "mu lo bakni" ==
> > > "mu bakni" == "five cows), works pretty
> much as before: "five
> > > things that really are cows".
> >
> > Was the "really are" intended?
> >
> > >From what I understand, "lo" is no longer
> makes a statement that
> > >a {lo
> > broda} "really is" a broda (or "really does"
> broda)... so wouldn't
> > {mu lo bakni} be closer to "five things that
> are cows", without
> > the "really"?
>
> You're conflating {mu lo bakni} with {lo
> bakni}. The former has an
> outer quantifier, the latter does not. The
> meanings are not the
> same. It is assumed that you wouldn't talk
> about a number of cows
> unless you actually wished to talk about cows.

But you might want to talk about a number (5 in
this case) of things that are related to cows
somehow, i.e., to specify the number of lo bakni
of current interest.

> > > * A fractional outer quantifier selects a
> portion of the
> > > group. So "pa pi re loi xa bakni cu bevri
> lo pipno" means that
> > > one half of some group of six cows (i.e. 3
> cows) carried the
> > > piano.
> >
> > Does {pa pi re} mean "one half", then?
>
> No, that was a mistake. I meant "pi mu".
>
> -Robin
>
>
>



Robin Lee Powell scripsit:

> > li pira'epavorebimuze du li fi'uze.
>
> No, actually. One is a number, the other is a fraction that
> represents that number, *sort* of. "li fi'u ze" is "one over seven";
> the fraction you gave is "li pa fe'i ze". To what extent these
> things are identical is unclear. Certainly they are different
> words, at the least.

They are different *words*, but they represent the same *entity*, which
is what "du" is all about. la kikeros. du la tulis. (Cicero is Tully)
not because the names are the same, but because the referent is the
same: Marcus Tullius Cicero, the great Roman orator.

--
Don't be so humble. You're not that great. John Cowan
--Golda Meir jcowan@reutershealth.com


posts: 2388



> Robin Lee Powell scripsit:
>
> > > li pira'epavorebimuze du li fi'uze.
> >
> > No, actually. One is a number, the other is a
> fraction that
> > represents that number, *sort* of. "li fi'u
> ze" is "one over seven";
> > the fraction you gave is "li pa fe'i ze". To
> what extent these
> > things are identical is unclear. Certainly
> they are different
> > words, at the least.
>
> They are different *words*, but they represent
> the same *entity*, which
> is what "du" is all about. la kikeros. du la
> tulis. (Cicero is Tully)
> not because the names are the same, but because
> the referent is the
> same: Marcus Tullius Cicero, the great Roman
> orator.
>
{kikeron} and {tulius} — native usage trumps
English habits.



posts: 14214

On Mon, Dec 27, 2004 at 06:59:31AM -0800, John E Clifford wrote:
>
> --- Robin Lee Powell
> <rlpowell@digitalkingdom.org> wrote:
>
> > On Sun, Dec 26, 2004 at 04:03:32PM +0100,
> > Philip Newton wrote:
> > > On Sat, 25 Dec 2004 12:38:17 -0800,
> > webmaster@lojban.org
> > > <webmaster@lojban.org> wrote:
> > > > Something that needs to be noted in
> > general: we, the BPFK, made
> > > > a consensus decision that we do not
> > make rulings on
> > > > ontological or metaphysical issues; that
> > is, we will not tell
> > > > you whether phrase X has meaning or
> > validity.
> > > ...
> > > > * lo with an outer quantifier, which is
> > exactly the same thing
> > > > as just sticking a number before an item
> > (i.e. "mu lo bakni" ==
> > > > "mu bakni" == "five cows), works pretty
> > much as before: "five
> > > > things that really are cows".
> > >
> > > Was the "really are" intended?
> > >
> > > >From what I understand, "lo" is no longer
> > makes a statement that
> > > >a {lo
> > > broda} "really is" a broda (or "really does"
> > broda)... so wouldn't
> > > {mu lo bakni} be closer to "five things that
> > are cows", without
> > > the "really"?
> >
> > You're conflating {mu lo bakni} with {lo
> > bakni}. The former has an
> > outer quantifier, the latter does not. The
> > meanings are not the
> > same. It is assumed that you wouldn't talk
> > about a number of cows
> > unless you actually wished to talk about cows.
>
> But you might want to talk about a number (5 in
> this case) of things that are related to cows
> somehow, i.e., to specify the number of lo bakni
> of current interest.

That's {lo mu bakni}.

-Robin


posts: 2388


<rlpowell@digitalkingdom.org> wrote:

> On Mon, Dec 27, 2004 at 06:59:31AM -0800, John
> E Clifford wrote:
> >
> > --- Robin Lee Powell
> > <rlpowell@digitalkingdom.org> wrote:
> >
> > > On Sun, Dec 26, 2004 at 04:03:32PM +0100,
> > > Philip Newton wrote:
> > > > On Sat, 25 Dec 2004 12:38:17 -0800,
> > > webmaster@lojban.org
> > > > <webmaster@lojban.org> wrote:
> > > > > Something that needs to be noted in
> > > general: we, the BPFK, made
> > > > > a consensus decision that we do
> not
> > > make rulings on
> > > > > ontological or metaphysical issues;
> that
> > > is, we will not tell
> > > > > you whether phrase X has meaning or
> > > validity.
> > > > ...
> > > > > * lo with an outer quantifier, which is
> > > exactly the same thing
> > > > > as just sticking a number before an
> item
> > > (i.e. "mu lo bakni" ==
> > > > > "mu bakni" == "five cows), works pretty
> > > much as before: "five
> > > > > things that really are cows".
> > > >
> > > > Was the "really are" intended?
> > > >
> > > > >From what I understand, "lo" is no
> longer
> > > makes a statement that
> > > > >a {lo
> > > > broda} "really is" a broda (or "really
> does"
> > > broda)... so wouldn't
> > > > {mu lo bakni} be closer to "five things
> that
> > > are cows", without
> > > > the "really"?
> > >
> > > You're conflating {mu lo bakni} with {lo
> > > bakni}. The former has an
> > > outer quantifier, the latter does not. The
> > > meanings are not the
> > > same. It is assumed that you wouldn't talk
> > > about a number of cows
> > > unless you actually wished to talk about
> cows.
> >
> > But you might want to talk about a number (5
> in
> > this case) of things that are related to cows
> > somehow, i.e., to specify the number of lo
> bakni
> > of current interest.
>
> That's {lo mu bakni}.

Isn't that some group of groups of five things
related to cows?
How do you say "five of the things just referred
to as {lo bakni}"?


posts: 1912


> --- Robin Lee Powell wrote:
> >
> > That's {lo mu bakni}.
>
> Isn't that some group of groups of five things
> related to cows?
> How do you say "five of the things just referred
> to as {lo bakni}"?

It depends what you want to say.

If you want to refer to five of the things just referred
to, then {lo mu lo bakni} for example would serve. If you
don't want to refer to them, but you want to say that exactly
five of those just referred to do something or other, then
{mu lo bakni} would be appropriate.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


posts: 2388


wrote:

>
> --- John E Clifford wrote:
> > --- Robin Lee Powell wrote:
> > >
> > > That's {lo mu bakni}.
> >
> > Isn't that some group of groups of five
> things
> > related to cows?
> > How do you say "five of the things just
> referred
> > to as {lo bakni}"?
>
> It depends what you want to say.
>
> If you want to refer to five of the things just
> referred
> to, then {lo mu lo bakni} for example would
> serve. If you
> don't want to refer to them, but you want to
> say that exactly
> five of those just referred to do something or
> other, then
> {mu lo bakni} would be appropriate.

I am afraid I don't quite follow what is going on
here. I gather that you want "refer" to have
some special sense, involving perhaps
identification or specification or whatever.
Could you give an example of the difference. We
have some sentence involving {lo bakni},
referring to some group of things related to
cows. I want to say that five of those things do
such and such. And have said that, I want to go
on and say some more things about those five. Do
I use {mu lo bakni} both times or {lo mu lo
bakni} both times or the short form the first
time and the long form the second (or, I suppose, conversely)?


posts: 1912


> I am afraid I don't quite follow what is going on
> here. I gather that you want "refer" to have
> some special sense, involving perhaps
> identification or specification or whatever.

No, just the usual sense as far as I can tell.
Terms that start with a gadri have referents.
Quantifiers are bridi operators that say how many
of the referents of the term they quantify satisfy
the bridi the term is in.

> Could you give an example of the difference. We
> have some sentence involving {lo bakni},
> referring to some group of things related to
> cows.

_Related_ to cows? {lo bakni} refers to cows,
not to things related to cows.

> I want to say that five of those things do
> such and such.

ko'a goi lo mu lo bakni cu broda

> And have said that, I want to go
> on and say some more things about those five.

ko'a brode

> Do
> I use {mu lo bakni} both times or {lo mu lo
> bakni} both times

The latter, preferrably linking them with a pronoun
if it's the same five cows both times.

> or the short form the first
> time and the long form the second

You could do that. You would not be referring to those five
cows the first time, you would just be saying exactly how
many out of the lot do something or other. Then the second
time pragmatically one would assume that the five cows you
are talking about are the ones that make the first sentence
true.

> (or, I suppose, conversely)?

That would not have the same pragmatic implication, I would
say.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


posts: 381

In a message dated 2004-12-27 9:20:00 AM Eastern Standard Time,
jcowan@reutershealth.com writes:


> In a context of exact numbers, no. Where measurements are in question, .5
> suggests an accuracy of one decimal place, whereas 1/2 suggests perfect
> (unattainable) accuracy.
>
> —
> John Cowan

I interpret 1/2 as being to the nearest half, so its precision is 1/5 that of
0.5.

stevo

posts: 2388


wrote:

>
> --- John E Clifford wrote:
> > I am afraid I don't quite follow what is
> going on
> > here. I gather that you want "refer" to have
> > some special sense, involving perhaps
> > identification or specification or whatever.
>
> No, just the usual sense as far as I can tell.
> Terms that start with a gadri have referents.
> Quantifiers are bridi operators that say how
> many
> of the referents of the term they quantify
> satisfy
> the bridi the term is in.

You said
<<If you want to refer to five of the things just
referred
to, then {lo mu lo bakni} for example would
serve. If you
don't want to refer to them, but you want to say
that exactly
five of those just referred to do something or
other, then
{mu lo bakni} would be appropriate.>>

The difference between referring to five of the
things just referred to and saying that five of
the things just referred to do something or other
is the distinction I am trying to get a grip on.
I don't know how one says that five of the things
do something or other without referring to them
nor can I see any likely point in referring to
five of these guys without saying that they do
(or are) something or other.

> > Could you give an example of the difference.
> We
> > have some sentence involving {lo bakni},
> > referring to some group of things related to
> > cows.
>
> _Related_ to cows? {lo bakni} refers to cows,
> not to things related to cows.

I am just going by the lesson which is meant to
explain what the Hell "generic reference" is all
about. I see that that example was changed in the
latest version of that page, thus making the
notion of generic reference again obscure but
less hopelessly stupid.

> > I want to say that five of those things do
> > such and such.
>
> ko'a goi lo mu lo bakni cu broda
>
> > And have said that, I want to go
> > on and say some more things about those five.
>
>
> ko'a brode

Well, the interesting case would be one that used
a full form rather than a pronoun. Suppose
that, when I said {lo mu lo bakni cu broda} I had
not decided to say more about them and so had not

{ko'a}d. What would {mu lo bakni cu broda} say?

> > Do
> > I use {mu lo bakni} both times or {lo mu lo
> > bakni} both times
>
> The latter, preferrably linking them with a
> pronoun
> if it's the same five cows both times.
>
> > or the short form the first
> > time and the long form the second
>
> You could do that. You would not be referring
> to those five
> cows the first time, you would just be saying
> exactly how
> many out of the lot do something or other. Then
> the second
> time pragmatically one would assume that the
> five cows you
> are talking about are the ones that make the
> first sentence
> true.

AHAH. {mu lo bakni} is not a reference (as in
older Lojban) to a bunch of five cows drawn from
the bunch referred to by {lo bakni} but something
less somehow — it tells howmany there are and
what they do but without referring to them. How
should the second reference to (well, the first

  • reference* but the second indication of)these

five cows be expressed in full form? I find this
reading of {lo} perverse — but you know that.

> > (or, I suppose, conversely)?
>
> That would not have the same pragmatic
> implication, I would
> say.



posts: 1912


> AHAH. {mu lo bakni} is not a reference (as in
> older Lojban) to a bunch of five cows drawn from
> the bunch referred to by {lo bakni} but something
> less somehow — it tells howmany there are and
> what they do but without referring to them.

That's right, but that is not a change. Quantifiers
in logic are bridi operators, they don't refer, and
quantifiers in Lojban have always been the usual
quantifiers of logic.

PA da zo'u da broda

says: the sentence "x broda" is satisfied by exactly PA things.
It makes no reference to the things that satisfy the sentence,
it only says exactly how many they are.

> How
> should the second reference to (well, the first
> *reference* but the second indication of)these
> five cows be expressed in full form?

With {lo} in front you have a reference.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


posts: 2388


wrote:

>
> --- John E Clifford wrote:
> > AHAH. {mu lo bakni} is not a reference (as
> in
> > older Lojban) to a bunch of five cows drawn
> from
> > the bunch referred to by {lo bakni} but
> something
> > less somehow — it tells howmany there are
> and
> > what they do but without referring to them.
>
> That's right, but that is not a change.
> Quantifiers
> in logic are bridi operators, they don't refer,
> and
> quantifiers in Lojban have always been the
> usual
> quantifiers of logic.
>
> PA da zo'u da broda
>
> says: the sentence "x broda" is satisfied by
> exactly PA things.
> It makes no reference to the things that
> satisfy the sentence,
> it only says exactly how many they are.

That is of course a quantifier on a variable, not
on a description. Are you suddenly going to be
modular and insist that a quantifier in one
context always behaves just like a quantifier in
another (it won't even work on your system of
course)?
> > How
> > should the second reference to (well, the
> first
> > *reference* but the second indication
> of)these
> > five cows be expressed in full form?
>
> With {lo} in front you have a reference.
>

I think I prefer the old way and still need to
see some reason for changing. It can't be
practicality nor modularity nor consistency, so
what is it that is so pressing as to change 50
years of work?
I note again, of course, that none of this is in
the offical definitions as given (unless they
have changed recently).



On Mon, 27 Dec 2004 18:05:10 EST, MorphemeAddict@wmconnect.com
<MorphemeAddict@wmconnect.com> wrote:
> In a message dated 2004-12-27 9:20:00 AM Eastern Standard Time,
> jcowan@reutershealth.com writes:
>
> > In a context of exact numbers, no. Where measurements are in question, .5
> > suggests an accuracy of one decimal place, whereas 1/2 suggests perfect
> > (unattainable) accuracy.
>
> I interpret 1/2 as being to the nearest half, so its precision is 1/5 that
> of 0.5.

Same — I don't interpret "8½×11" paper as being 8.50000000 by
11.000000 inches, for example. Or a baby being born which is described
as "20½ inches long" as having a length of 20.50 inches — more like
20.5. (And possibly with an error of 0.125 in either way, rather than
merely 0.05 in which I might expect had the weight been given in
decimal numbers.)

Cheers,
--

Philip Newton <philip.newton@gmail.com>



posts: 1912

> That is of course a quantifier on a variable, not
> on a description. Are you suddenly going to be
> modular and insist that a quantifier in one
> context always behaves just like a quantifier in
> another (it won't even work on your system of
> course)?

PA sumti = PA da poi ke'a me sumti

> I note again, of course, that none of this is in
> the offical definitions as given (unless they
> have changed recently).

It's been there for months, and you have commented
profusely on it, so you can't say you haven't seen it
before.

mu'o mi'e xorxes





__
Do you Yahoo!?
Send holiday email and support a worthy cause. Do good.
http://celebrity.mail.yahoo.com


posts: 2388

wrote:

> --- John E Clifford wrote:
> > That is of course a quantifier on a variable,
> not
> > on a description. Are you suddenly going to
> be
> > modular and insist that a quantifier in one
> > context always behaves just like a quantifier
> in
> > another (it won't even work on your system of
> > course)?
>
> PA sumti = PA da poi ke'a me sumti

So you say. I see no reason to think this is
correct (never mind worrying about circularity.

>
> > I note again, of course, that none of this is
> in
> > the offical definitions as given (unless they
> > have changed recently).
>
> It's been there for months, and you have
> commented
> profusely on it, so you can't say you haven't
> seen it
> before.

I meant in either of the pieces called
definitions. The add-ons are another matter entirely.


posts: 1912


> > PA sumti = PA da poi ke'a me sumti
>
> So you say. I see no reason to think this is
> correct

It's a definition.

> (never mind worrying about circularity.

It's not circular. You can trust me about that, or you can
check it out for yourself. (Or else show us how it is circular.)

> I meant in either of the pieces called
> definitions. The add-ons are another matter entirely.

The definition in English says:

"An outer quantifier can be used to quantify distributively
over such individuals."

The formal definition is the one I wrote above, and says
the same thing.

mu'o mi'e xorxes




__
Do you Yahoo!?
Yahoo! Mail - Helps protect you from nasty viruses.
http://promotions.yahoo.com/new_mail


posts: 2388

wrote:

> --- John E Clifford wrote:
>
> > > PA sumti = PA da poi ke'a me sumti
> >
> > So you say. I see no reason to think this is
> > correct
>
> It's a definition.

A definition for xorlan; I meant Lojban (as you
know).



> > I meant in either of the pieces called
> > definitions. The add-ons are another matter
> entirely.
>
> The definition in English says:
>
> "An outer quantifier can be used to quantify
> distributively
> over such individuals."
>
> The formal definition is the one I wrote above,
> and says
> the same thing.
>
As you know, I meant the definitions of {lo},
from which none of this follows nor with which
none is demonstrated compatible (although, since
taken together they are contradictory, everything follows).


posts: 2388

Still working on the strange use of "refer" that
seems to play some role here, is the following
reasnably correct:

If I say {mu lo bakni} twice, the sentences
involved might be made true by two (at least
partially) different bunches of cows. In neither
case are the cows referred to.

If I want to be sure that it is the same bunch of
cows the second time, I should use {lo mu lo
bakni} which assures that the reference in the
second case is to the cows that made the first
case true (but which were not there referred to).

If I use {lo mu lo bakni} in the first case I not
only claim that there are exactly five cows that
make the sentence involved true but I also refer
to them.

I am still unclear about why {mu lo bakni} does
not refer to the the cows. I am not even clear
why {Q da} doesn't rfer to what makes the
sentence true, labeit without giving a lot of
information about them — but that is not
required for reference. I suppose this is just
definitional at some level, but it seems to be
making a difference and to be somehow involved a
justification for some of the innovations that
you are foisting on Lojban.


posts: 2388

wrote:

> Still working on the strange use of "refer"
> that
> seems to play some role here, is the following
> reasnably correct:
>
> If I say {mu lo bakni} twice, the sentences
> involved might be made true by two (at least
> partially) different bunches of cows. In
> neither
> case are the cows referred to.
>
> If I want to be sure that it is the same bunch
> of
> cows the second time, I should use {lo mu lo
> bakni} which assures that the reference in the
> second case is to the cows that made the first
> case true (but which were not there referred
> to).
>
> If I use {lo mu lo bakni} in the first case I
> not
> only claim that there are exactly five cows
> that
> make the sentence involved true but I also
> refer
> to them.
>
> I am still unclear about why {mu lo bakni} does
> not refer to the the cows. I am not even clear
> why {Q da} doesn't refer to what makes the
> sentence true, labeit without giving a lot of
> information about them — but that is not
> required for reference. I suppose this is just
> definitional at some level, but it seems to be
> making a difference and to be somehow involved
> a
> justification for some of the innovations that
> you are foisting on Lojban.

<<hit the wrong damned button!>>

In my mind the fact that I can meaningfully talk
about picking out even pragmatically the the five
cows that made the earlier sentence true means
that they have already been introduced into the
context and I am unclear how that is done if not
by reference: they are the values of the
variables (if you insist that there are variables
in this case) in the first reference as they are
the values of {lo mu lo bakni} in the second
(Actually, of course, it is the bunch of them
that is the value in each case, or if you want to
insist that {mu lo bakni} is really {mu da poi
bakni}, quite independent of whatever {lo bakni}
refers to, the the second refers to the bunch of
those cows referred to eventually in the first.I
assume we are not yet doing plural
interpretations)
As I said somehwere earlier, we are constructing
the model as we go here and so what we bring with
sumti of any sort go into the model as the
reality end of the reference function. If I can
tell that there are five, I can tell which ones
they are in a sufficient way — though maybe in
only a rather a rather vague. I can, however,
know whether it is the same five involved at the
next stage.


posts: 1912

> --- John E Clifford <clifford-j@sbcglobal.net>
> wrote:
> > Still working on the strange use of "refer"
> > that
> > seems to play some role here, is the following
> > reasnably correct:

(I will answer assuming we are now talking of the
definitions approved by the BPFK.)

> > If I say {mu lo bakni} twice, the sentences
> > involved might be made true by two (at least
> > partially) different bunches of cows. In
> > neither
> > case are the cows referred to.

Right. For example:

mu lo bakni cu blabi ije mu lo bakni cu ca'o citka
Five cows are white, and five cows are eating.

I am not saying how many cows (if any) are both white
and eating. It could be anywhere from zero to five.

> > If I want to be sure that it is the same bunch
> > of
> > cows the second time, I should use {lo mu lo
> > bakni} which assures that the reference in the
> > second case is to the cows that made the first
> > case true (but which were not there referred
> > to).

No assurance, no. It may be the most likely assumption,
but that would depend on context.

> > If I use {lo mu lo bakni} in the first case I
> > not
> > only claim that there are exactly five cows
> > that
> > make the sentence involved true but I also
> > refer
> > to them.

No, you don't claim that there are exactly five cows
that make the sentence true, there may be more. You
refer to exactly five cows, and claim, of those, that
they make the sentence true.

> > I am still unclear about why {mu lo bakni} does
> > not refer to the the cows.

Because quantifiers are bridi operators, they don't
create a referring term.

> In my mind the fact that I can meaningfully talk
> about picking out even pragmatically the the five
> cows that made the earlier sentence true means
> that they have already been introduced into the
> context and I am unclear how that is done if not
> by reference:

Because you don't need to pick anything to claim that
five do something. The claim is meaningful without
any reference going on.

> they are the values of the
> variables (if you insist that there are variables
> in this case) in the first reference as they are
> the values of {lo mu lo bakni} in the second

The variables take _all_ the values of their range,
not just those values that make the sentence true.

> As I said somehwere earlier, we are constructing
> the model as we go here and so what we bring with
> sumti of any sort go into the model as the
> reality end of the reference function.

Yes, {lo bakni} in {mu lo bakni} does refer. Possibly
and probably to more than five cows. All of those
referents (not just the five that make the sentence true)
do go into the model.

> If I can
> tell that there are five, I can tell which ones
> they are in a sufficient way — though maybe in
> only a rather a rather vague. I can, however,
> know whether it is the same five involved at the
> next stage.

When you want to do reference, you can.

mu'o mi'e xorxes




__
Do you Yahoo!?
Meet the all-new My Yahoo! - Try it today!
http://my.yahoo.com




posts: 2388

wrote:

> --- John E Clifford wrote:
> > --- John E Clifford
> <clifford-j@sbcglobal.net>
> > wrote:
> > > Still working on the strange use of "refer"
> > > that
> > > seems to play some role here, is the
> following
> > > reasnably correct:
>
> (I will answer assuming we are now talking of
> the
> definitions approved by the BPFK.)
>
> > > If I say {mu lo bakni} twice, the sentences
> > > involved might be made true by two (at
> least
> > > partially) different bunches of cows. In
> > > neither
> > > case are the cows referred to.
>
> Right. For example:
>
> mu lo bakni cu blabi ije mu lo bakni cu ca'o
> citka
> Five cows are white, and five cows are
> eating.
>
> I am not saying how many cows (if any) are both
> white
> and eating. It could be anywhere from zero to
> five.
>
> > > If I want to be sure that it is the same
> bunch
> > > of
> > > cows the second time, I should use {lo mu
> lo
> > > bakni} which assures that the reference in
> the
> > > second case is to the cows that made the
> first
> > > case true (but which were not there
> referred
> > > to).
>
> No assurance, no. It may be the most likely
> assumption,
> but that would depend on context.

How do I get assurance? I can't use a pronoun
since it only works if the reference is the same,
but I have no reference yet.

>
> > > If I use {lo mu lo bakni} in the first case
> I
> > > not
> > > only claim that there are exactly five cows
> > > that
> > > make the sentence involved true but I also
> > > refer
> > > to them.
>
> No, you don't claim that there are exactly five
> cows
> that make the sentence true, there may be more.
> You
> refer to exactly five cows, and claim, of
> those, that
> they make the sentence true.

Good, we do agree on something.

> > > I am still unclear about why {mu lo bakni}
> does
> > > not refer to the the cows.
>
> Because quantifiers are bridi operators, they
> don't
> create a referring term.

Well, here is my problem: I believe that {lo
bakni} for example is both a referring expression
and a quantifier. Its being a quantifier — and
a particuilar one at that — is the major part of
what it means — for me — to say it is generic
or non-specific. I don't quite understand what
"bridi operator" means here that keeps it from
making referring expressions. After all, the
descriptor (even if you don't think it is a
quantifer) is a bridi operator in the normal
sense of the phrase and you hold it creates a
referring expression. Is it the difference
between a (to put this in Lojban terms)
bridi-forming operator and a sumti-forming
operator? If that is the case, then I will accept
that it is the whole bridi which makes the
reference, not just the term (though I don't
really believe this).

> > In my mind the fact that I can meaningfully
> talk
> > about picking out even pragmatically the the
> five
> > cows that made the earlier sentence true
> means
> > that they have already been introduced into
> the
> > context and I am unclear how that is done if
> not
> > by reference:
>
> Because you don't need to pick anything to
> claim that
> five do something. The claim is meaningful
> without
> any reference going on.

I don't quite see how but then I am now perfectly
sure that I don't understand your use of "refer."
I wonder if there is some other terminology we
could use that would make my point and eventually
make your point clear.

> > they are the values of the
> > variables (if you insist that there are
> variables
> > in this case) in the first reference as they
> are
> > the values of {lo mu lo bakni} in the second
>
> The variables take _all_ the values of their
> range,
> not just those values that make the sentence
> true.

We are not talking about the reange but the
values here. It is the values that are alluded
to in a
> > As I said somehwere earlier, we are
> constructing
> > the model as we go here and so what we bring
> with
> > sumti of any sort go into the model as the
> > reality end of the reference function.
>
> Yes, {lo bakni} in {mu lo bakni} does refer.
> Possibly
> and probably to more than five cows. All of
> those
> referents (not just the five that make the
> sentence true)
> do go into the model.
>
> > If I can
> > tell that there are five, I can tell which
> ones
> > they are in a sufficient way — though maybe
> in
> > only a rather a rather vague. I can,
> however,
> > know whether it is the same five involved at
> the
> > next stage.
>
> When you want to do reference, you can.
>
> mu'o mi'e xorxes
>
>
>
>
> __
> Do you Yahoo!?
> Meet the all-new My Yahoo! - Try it today!
> http://my.yahoo.com
>
>
>
>
>



posts: 2388

wrote:

> --- Jorge Llambías
> <jjllambias2000@yahoo.com.ar>
> wrote:
>
> > --- John E Clifford wrote:
> > > --- John E Clifford
> > <clifford-j@sbcglobal.net>
> > > wrote:
> > > > Still working on the strange use of
> "refer"
> > > > that
> > > > seems to play some role here, is the
> > following
> > > > reasnably correct:
> >
> > (I will answer assuming we are now talking of
> > the
> > definitions approved by the BPFK.)
> >
> > > > If I say {mu lo bakni} twice, the
> sentences
> > > > involved might be made true by two (at
> > least
> > > > partially) different bunches of cows. In
> > > > neither
> > > > case are the cows referred to.
> >
> > Right. For example:
> >
> > mu lo bakni cu blabi ije mu lo bakni cu
> ca'o
> > citka
> > Five cows are white, and five cows are
> > eating.
> >
> > I am not saying how many cows (if any) are
> both
> > white
> > and eating. It could be anywhere from zero to
> > five.
> >
> > > > If I want to be sure that it is the same
> > bunch
> > > > of
> > > > cows the second time, I should use {lo mu
> > lo
> > > > bakni} which assures that the reference
> in
> > the
> > > > second case is to the cows that made the
> > first
> > > > case true (but which were not there
> > referred
> > > > to).
> >
> > No assurance, no. It may be the most likely
> > assumption,
> > but that would depend on context.
>
> How do I get assurance? I can't use a pronoun
> since it only works if the reference is the
> same,
> but I have no reference yet.
>
> >
> > > > If I use {lo mu lo bakni} in the first
> case
> > I
> > > > not
> > > > only claim that there are exactly five
> cows
> > > > that
> > > > make the sentence involved true but I
> also
> > > > refer
> > > > to them.
> >
> > No, you don't claim that there are exactly
> five
> > cows
> > that make the sentence true, there may be
> more.
> > You
> > refer to exactly five cows, and claim, of
> > those, that
> > they make the sentence true.
>
> Good, we do agree on something.
>
> > > > I am still unclear about why {mu lo
> bakni}
> > does
> > > > not refer to the the cows.
> >
> > Because quantifiers are bridi operators, they
> > don't
> > create a referring term.
>
> Well, here is my problem: I believe that {lo
> bakni} for example is both a referring
> expression
> and a quantifier. Its being a quantifier --
> and
> a particuilar one at that — is the major part
> of
> what it means — for me — to say it is generic
> or non-specific. I don't quite understand what
> "bridi operator" means here that keeps it from
> making referring expressions. After all, the
> descriptor (even if you don't think it is a
> quantifer) is a bridi operator in the normal
> sense of the phrase and you hold it creates a
> referring expression. Is it the difference
> between a (to put this in Lojban terms)
> bridi-forming operator and a sumti-forming
> operator? If that is the case, then I will
> accept
> that it is the whole bridi which makes the
> reference, not just the term (though I don't
> really believe this).
>
> > > In my mind the fact that I can meaningfully
> > talk
> > > about picking out even pragmatically the
> the
> > five
> > > cows that made the earlier sentence true
> > means
> > > that they have already been introduced into
> > the
> > > context and I am unclear how that is done
> if
> > not
> > > by reference:
> >
> > Because you don't need to pick anything to
> > claim that
> > five do something. The claim is meaningful
> > without
> > any reference going on.
>
> I don't quite see how but then I am now
> perfectly
> sure that I don't understand your use of
> "refer."
> I wonder if there is some other terminology we
> could use that would make my point and
> eventually
> make your point clear.
>
> > > they are the values of the
> > > variables (if you insist that there are
> > variables
> > > in this case) in the first reference as
> they
> > are
> > > the values of {lo mu lo bakni} in the
> second
> >
> > The variables take _all_ the values of their
> > range,
> > not just those values that make the sentence
> > true.
>
> We are not talking about the range but the
> values here. It is the values that are alluded
> to in a quantified statement, they are what
makes the sentence true (damned button, whatever
one it is!)
> > > As I said somehwere earlier, we are
> > constructing
> > > the model as we go here and so what we
> bring
> > with
> > > sumti of any sort go into the model as the
> > > reality end of the reference function.
> >
> > Yes, {lo bakni} in {mu lo bakni} does refer.
> > Possibly
> > and probably to more than five cows. All of
> > those
> > referents (not just the five that make the
> > sentence true)
> > do go into the model.

Hey two things we agree on, but this seems
inconsistent with what you have said elsewhere
which appear to say that {mu lo bakni} = {mu
bakni} quantifies over cows, not just over lo
bakni.

> > > If I can
> > > tell that there are five, I can tell which
> > ones
> > > they are in a sufficient way — though
> maybe
> > in
> > > only a rather a rather vague. I can,
> > however,
> > > know whether it is the same five involved
> at
> > the
> > > next stage.
> >
> > When you want to do reference, you can.

I already did in some standard sense of
"reference." Yours may be a standard sense as
well but I don't know it nor how it goes.




posts: 1912

> > > Yes, {lo bakni} in {mu lo bakni} does refer.
> > > Possibly
> > > and probably to more than five cows. All of
> > > those
> > > referents (not just the five that make the
> > > sentence true)
> > > do go into the model.
>
> Hey two things we agree on, but this seems
> inconsistent with what you have said elsewhere
> which appear to say that {mu lo bakni} = {mu
> bakni} quantifies over cows, not just over lo
> bakni.

{mu bakni} does quantify over cows, and cows is perecisely
what {lo bakni} refers to, so they are the same thing.
{lo bakni} refers to the cows in the model, as you put it,
and both {mu bakni} and {mu lo bakni} quantify over
the cows in the model.

mu'o mi'e xorxes




__
Do you Yahoo!?
Dress up your holiday email, Hollywood style. Learn more.
http://celebrity.mail.yahoo.com


posts: 2388

wrote:

> --- John E Clifford wrote:
> > > > Yes, {lo bakni} in {mu lo bakni} does
> refer.
> > > > Possibly
> > > > and probably to more than five cows. All
> of
> > > > those
> > > > referents (not just the five that make
> the
> > > > sentence true)
> > > > do go into the model.
> >
> > Hey two things we agree on, but this seems
> > inconsistent with what you have said
> elsewhere
> > which appear to say that {mu lo bakni} = {mu
> > bakni} quantifies over cows, not just over lo
> > bakni.
>
> {mu bakni} does quantify over cows, and cows is
> perecisely
> what {lo bakni} refers to, so they are the same
> thing.
> {lo bakni} refers to the cows in the model, as
> you put it,
> and both {mu bakni} and {mu lo bakni} quantify
> over
> the cows in the model.

While I suspect you of being deliberately
obfuscatory, I'll concede that I was not as
precise as apparently I needed to be, assuming
that some things were in our common (assumed
cooperative) context. So, what I meant was
"You say here that the quantifier in {mu lo
bakni}has {lo bakni} as its range but other
places it appears that {mu lo bakni} = {mu bakni}
has the whole of existing cows as its range.
Which is correct?"


posts: 1912

> So, what I meant was
> "You say here that the quantifier in {mu lo
> bakni}has {lo bakni} as its range but other
> places it appears that {mu lo bakni} = {mu bakni}
> has the whole of existing cows as its range.
> Which is correct?"

{mu lo bakni} = {mu bakni} has {lo bakni} as its range,
i.e. the whole of cows in the model.

"Existing cows", as in the cows that exist (or ever existed
or ever will exist?) in the real world is one particular model
that does not always apply. It will apply in some contexts.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


posts: 2388

wrote:

> --- John E Clifford wrote:
> > So, what I meant was
> > "You say here that the quantifier in {mu lo
> > bakni}has {lo bakni} as its range but other
> > places it appears that {mu lo bakni} = {mu
> bakni}
> > has the whole of existing cows as its range.
> > Which is correct?"
>
> {mu lo bakni} = {mu bakni} has {lo bakni} as
> its range,
> i.e. the whole of cows in the model.
>
> "Existing cows", as in the cows that exist (or
> ever existed
> or ever will exist?) in the real world is one
> particular model
> that does not always apply. It will apply in
> some contexts.

"in the model" still puzzles me. All the cows
that .... what? I am thinking about?, are near
enough to be of interest? and so on. I would
have thought (because that is the way it has
been) that {lo bakni} was always a selection from
available cows, that is that the range on a
quantifier might well be greater than {lo bakni}
(unless it was lo ro bakni), though it would
generally be less than all the cows that are,
were or ever will be (usually just all that
there are now or even, as noted, all that are
currently available, not all of which need be in
lo bakni).

On a not unrelated issue. Two occurrences of {mu
bakni} can indicate two different bunches of
cows; can two occurrences of {lo bakni} (in the
same context, etc.) indicate two different
bunches of cows or must they be the same?
Gneralizinf various other things it sems that a
good case could be made either way.


posts: 1912

> "in the model" still puzzles me.

You introduced the term, not me. I was using
"universe of discourse".

> All the cows
> that .... what? I am thinking about?, are near
> enough to be of interest? and so on.

Right. All the ones we are talking about. All the
ones that are part of the discourse.

> On a not unrelated issue. Two occurrences of {mu
> bakni} can indicate two different bunches of
> cows;

While the universe of discourse does not change, each
occurrence says that out of the (same) available
cows, five do something or other. It need not be the same
five each time.

> can two occurrences of {lo bakni} (in the
> same context, etc.) indicate two different
> bunches of cows or must they be the same?

As long as the model doesn't change, they are the same,
but the model can change very quckly. As you said, the
model is being constructed and modified as discourse
progresses. There are ways (such as assignable pronouns)
to ensure more durable reference.

mu'o mi'e xorxes




__
Do you Yahoo!?
Yahoo! Mail - Helps protect you from nasty viruses.
http://promotions.yahoo.com/new_mail


posts: 2388

wrote:

> --- John E Clifford wrote:
> > "in the model" still puzzles me.
>
> You introduced the term, not me. I was using
> "universe of discourse".

OK, that translation helps. It's not quite what
I meant but we are getting to the same page.

> > All the cows
> > that .... what? I am thinking about?, are
> near
> > enough to be of interest? and so on.
>
> Right. All the ones we are talking about. All
> the
> ones that are part of the discourse.

{lo bakni} is all of these? At any given moment
at least? I suppose that will work if the whole
can change constantly. My universe of discourse
is relatively stable but what parts of it are
taken out to be referred to grows as the model
develops.
Thus using, say, {lo drata bakni} after {lo
bakni} would not add to the universe of discourse
but would add to what is used in the model.

> > On a not unrelated issue. Two occurrences of
> {mu
> > bakni} can indicate two different bunches of
> > cows;
>
> While the universe of discourse does not
> change, each
> occurrence says that out of the (same)
> available
> cows, five do something or other. It need not
> be the same
> five each time.

That is, another bunch might get taken up in the
model. OK

> > can two occurrences of {lo bakni} (in the
> > same context, etc.) indicate two different
> > bunches of cows or must they be the same?
>
> As long as the model doesn't change, they are
> the same,
> but the model can change very quckly. As you
> said, the
> model is being constructed and modified as
> discourse
> progresses. There are ways (such as assignable
> pronouns)
> to ensure more durable reference.

This one is harder to see. You mean, I suppose
either that the universe of discourse can expand
or that the parts taken up in the model can
expand. The latter is clearly true, the former
seems to equate the universe of discourse with
the things taken up in the model, which is fine
except for the metaphysical question of where
they came from. Then, of course, if {lo bakni}
is all the cows in the universe of discourse =
all those playing the role of a value in the
model, it will change if some new cows come in,
lo drata bakni, say (this generates a paradox of
a sort, requiring that we go back and change the
model we were developing, but realistically this
sort of thing happens — just not obviously in
this case.)




posts: 1912

> How do I get assurance? I can't use a pronoun
> since it only works if the reference is the same,
> but I have no reference yet.

How do you ever get assurance that your first reference to
something gets what you want?

If you say {mu bakni cu blabi} and then you want to refer to
the five cows that are white, you can use {lo mu bakni poi blabi}.
Or {lo mu bakni poi go'i} if the first bridi was longer and you
don't want to repeat it. In other words, to refer to the things
that satisfy a certain sentence you use "the things that satisfy
such sentence".

> I don't quite understand what
> "bridi operator" means here that keeps it from
> making referring expressions.

By bridi operator I meant an operator that takes
a bridi as an argument and returns a bridi. A function
from bridi to bridi.

> After all, the
> descriptor (even if you don't think it is a
> quantifer) is a bridi operator in the normal
> sense of the phrase and you hold it creates a
> referring expression.

A gadri is a function from bridi to sumti, yes.

> Is it the difference
> between a (to put this in Lojban terms)
> bridi-forming operator and a sumti-forming
> operator?

Right.

> If that is the case, then I will accept
> that it is the whole bridi which makes the
> reference, not just the term (though I don't
> really believe this).

I don't have a problem with that. But then you
need to turn that bridi into a sumti in order to
use it as a reference, and that's what gadri do.

mu'o mi'e xorxes




__
Do you Yahoo!?
Yahoo! Mail - Find what you need with new enhanced search.
http://info.mail.yahoo.com/mail_250


posts: 2388

wrote:

> --- John E Clifford wrote:

>
> By bridi operator I meant an operator that
> takes
> a bridi as an argument and returns a bridi. A
> function
> from bridi to bridi.
>
> > After all, the
> > descriptor (even if you don't think it is a
> > quantifer) is a bridi operator in the normal
> > sense of the phrase and you hold it creates a
> > referring expression.
>
> A gadri is a function from bridi to sumti, yes.
>
> > Is it the difference
> > between a (to put this in Lojban terms)
> > bridi-forming operator and a sumti-forming
> > operator?
>
> Right.
>
> > If that is the case, then I will accept
> > that it is the whole bridi which makes the
> > reference, not just the term (though I don't
> > really believe this).
>
> I don't have a problem with that. But then you
> need to turn that bridi into a sumti in order
> to
> use it as a reference, and that's what gadri
> do.
>
Puzzlement. I thought you just agreed that the
sentence as a whole makes the reference. What
more can using a gadri do?


posts: 1912

> > > If that is the case, then I will accept
> > > that it is the whole bridi which makes the
> > > reference, not just the term (though I don't
> > > really believe this).
> >
> > I don't have a problem with that. But then you
> > need to turn that bridi into a sumti in order
> > to
> > use it as a reference, and that's what gadri
> > do.
> >
> Puzzlement. I thought you just agreed that the
> sentence as a whole makes the reference. What
> more can using a gadri do?

It allows you to use the reference to make a claim about
its referents.

mu'o mi'e xorxes




__
Do you Yahoo!?
Send holiday email and support a worthy cause. Do good.
http://celebrity.mail.yahoo.com


posts: 2388

wrote:

> --- John E Clifford wrote:
> > > > If that is the case, then I will accept
> > > > that it is the whole bridi which makes
> the
> > > > reference, not just the term (though I
> don't
> > > > really believe this).
> > >
> > > I don't have a problem with that. But then
> you
> > > need to turn that bridi into a sumti in
> order
> > > to
> > > use it as a reference, and that's what
> gadri
> > > do.
> > >
> > Puzzlement. I thought you just agreed that
> the
> > sentence as a whole makes the reference.
> What
> > more can using a gadri do?
>
> It allows you to use the reference to make a
> claim about
> its referents.

But I have already done that in the first place;
even though it is the whole sentence which makes
the reference it does also make a claim about
that referent.


posts: 1912

> > What
> > > more can using a gadri do?
> >
> > It allows you to use the reference to make a
> > claim about
> > its referents.
>
> But I have already done that in the first place;
> even though it is the whole sentence which makes
> the reference it does also make a claim about
> that referent.

Consider these three cases:

(1) Exactly three cows are white.
(2) At least three cows are white.
(3) At most three cows are white.

With (1), I suppose you could argue that you made a reference
to all the cows that are white, and claimed about them that
they are white. I would not say that, I would say that all
one does is say how many white cows there are without referring
to the white cows.

With (2), the presumed referents are more doubtful. Are they
all the white cows, or any group of at least three white cows?
I would say that all one does with (2) is say how many white
cows there are without referring to the white cows.

With (3), we are not even sure that there are any white cows,
so talking of reference is even more iffy. I would say that
(3) just says how many white cows there are without referring
to the white cows.

True references to the white cows would be "the exactly three
cows that are white", "the at least three cows that are white"
and "the at most three cows that are white". In lojban:

lo ci bakni (poi blabi)
lo su'o ci bakni (poi blabi)
lo su'e ci bakni (poi blabi)


PA bakni cu blabi

says that the number of cows that are white is PA. It does
not refer to PA cows and say of those that they are white.
The latter is meaningless for some PA, and iffy for others.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250


posts: 2388

wrote:

> --- John E Clifford wrote:
> > > What
> > > > more can using a gadri do?
> > >
> > > It allows you to use the reference to make
> a
> > > claim about
> > > its referents.
> >
> > But I have already done that in the first
> place;
> > even though it is the whole sentence which
> makes
> > the reference it does also make a claim about
> > that referent.
>
> Consider these three cases:
>
> (1) Exactly three cows are white.
> (2) At least three cows are white.
> (3) At most three cows are white.
>
> With (1), I suppose you could argue that you
> made a reference
> to all the cows that are white, and claimed
> about them that
> they are white. I would not say that, I would
> say that all
> one does is say how many white cows there are
> without referring
> to the white cows.

I would, of course, like to submit here that what
I am talking about is {ci (lo) bakni cu blabi} or
some such so it is not "exactly three" in the
sense that none other are, but just "three" — as
you pointed out before. This does not make a lot
of difference but I try to keep the cases clear
in case it does.
> With (2), the presumed referents are more
> doubtful. Are they
> all the white cows, or any group of at least
> three white cows?
> I would say that all one does with (2) is say
> how many white
> cows there are without referring to the white
> cows.
>
> With (3), we are not even sure that there are
> any white cows,
> so talking of reference is even more iffy. I
> would say that
> (3) just says how many white cows there are
> without referring
> to the white cows.

The referring in each case is through the
particular quantifier (as it were), which is all
of these, though admittedly buried in very
different ways. All that the last case says is
that what is referred to may not be white. the
second says that we don't know how many are in
the group referred to — often the case even with
definite descriptions.


> True references to the white cows would be "the
> exactly three
> cows that are white", "the at least three cows
> that are white"
> and "the at most three cows that are white". In
> lojban:
>
> lo ci bakni (poi blabi)
> lo su'o ci bakni (poi blabi)
> lo su'e ci bakni (poi blabi)

These have some of the same problems as the cases
above — we don't know how many are involved in
the lastt two cases and in the last case the fact
that there may be none with the resultant
problems about claims that, taken grammatically,
are true but referentially false (Oh, I suppose
you are going to allow nonreferring sumti to
participate in true atomic sentences, but causes
a different set of problems, which I don't know
how you would deal with).

> PA bakni cu blabi
>
> says that the number of cows that are white is
> PA. It does
> not refer to PA cows and say of those that they
> are white.
> The latter is meaningless for some PA, and iffy
> for others.

Well, it is not exactly meaningless when PA =
{no}, it is just that its meaning does not flow
out in the usual pattern, which pattern says
exactly that reference does not occur in this
case. Otherwise they all get treated the same
way (with some tranlation work required for
proportional and relative quantifiers).

As McKay points out, descriptions are just
quantifiers (and, I suppose, conversely). That
is, a description is a function from a
sentence-matrix to a sentence as much as a
quantifier is. (This was always one of the
hardest pieces of Montague grammar to get one's
head around but it is very old in logic --
singular propositions are on a par with
universals and particulars, just different functions).


posts: 1912

> How about:
>
> In some discussions, saying "mi kalte pa lo pavyseljirna" (which
> litterally means "there exists one thing that is a unicorn that
> I am hunting"; this implies that at least one unicorn exists) is
> perfectly reasonable,

That's good.

> > For me, the most helpful concept in that article is the idea of
> > plural terms: terms that refer to several things without having to
> > create a new entity that comprises them (a "group" or "mass").
>
> <nod> Do we use that at all?

Yes: lo/le/la are all plural referers.

> > > "lo cribe" could be one, or a billion, or none (although expect
> > > listener hostility!!),
> >
> > What do you have in mind by saying that it can be none?
>
> "I need a doctor".

That cannot mean that I need no doctor, though.

mu'o mi'e xorxes




__
Do you Yahoo!?
Meet the all-new My Yahoo! - Try it today!
http://my.yahoo.com