User:Ciste/Talk:BPFK Section: gadri

From Lojban
Jump to navigation Jump to search

This template is used at the beginning of BPFK Section pages imported from the lojban Tiki.

This page is imported from the page from the lojban Tiki.


arj Posted by arj on Wed 19 of May, 2004 10:33 GMT posts: 953

I dislike the new semantics of the {lo} article. Quite a lot, actually.

Are there by the way any other changes to the other articles? A cursory read doesn't reveal any.

Mightn't Nick's gadri solution be the way to go?



Posted by Anonymous on Wed 19 of May, 2004 14:33 GMT

arj:

> Are there by the way any other changes to the other articles? A cursory read

> doesn't reveal any.

{le} and {la} become constants instead of being automatically

quantified as {ro le} and {ro la}.

The change is almost imperceptible in practice because

{le} and {la} are almost always used for singletons. When

used in the plural, however, the proposed definitions make

them easier to handle because with constants you don't need

to worry about scope. An explicit {ro} gives you back the

quantified forms.

lo'i/le'i/la'i/loi/lei/lai/lo'e/le'e I tried to leave

untouched. The official definitions are not very clear in

some respects however. For example, what is {pimu lo'i broda}?

That's something like "a set with half of all the broda there

are", but is it such a set generically, or is it existentially

quantified (i.e. "at least one set with half the cardinality of

the total set"), or what? This has seen no significant

usage, and it is probably not very useful in any case, but

it is ill defined. I'm not sure whether I should try to

come up with something more precise but which will deviate

from the traditional prescription. From my point of view

it is not worth it because these cmavo should be phased out,

but I will attempt it if there is a demand for it.

> Migtn't Nick's gadri solution be the way to go?

Which one? Adding a new cmavo for generic? That would have

a much bigger impact on past usage, wouldn't it?

mu'o mi'e xorxes

'__

Do you Yahoo!?

SBC Yahoo! - Internet access at a great low price.

http://promo.yahoo.com/sbc/


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Wed 19 of May, 2004 16:39 GMT posts: 14214

> arj:

> I dislike the new semantics of the {lo} article. Quite a lot, actually.

Can you explain your objections, please?

-Robin



Posted by Anonymous on Wed 19 of May, 2004 18:03 GMT

On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

>

> arj:

> > Are there by the way any other changes to the other articles? A

> > cursory read doesn't reveal any.

>

> {le} and {la} become constants instead of being automatically

> quantified as {ro le} and {ro la}.

What do you mean by 'constants'?

> I'm not sure whether I should try to come up with something more

> precise but which will deviate from the traditional prescription. From

> my point of view it is not worth it because these cmavo should be

> phased out, but I will attempt it if there is a demand for it.

I would like to go on record as both not disliking the cmavo in question

and preferring that as many things be well defined as possible.

Why do you dislike those other gadri so much, anyways?

-Robin



Posted by Anonymous on Wed 19 of May, 2004 19:05 GMT


> On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

> > {le} and {la} become constants instead of being automatically

> > quantified as {ro le} and {ro la}.

>

> What do you mean by 'constants'?

Constants are direct references, not quantification over some set.

With constants, you single out an individual (or group) and

give a relationship in which it participates.

With quantification you single out a set, and then you say

how the members of the set are distributed with respect to

a relationship.

In simple cases it makes little difference which way you go,

but when you have two or three quantifiers operating or

some other thing with scope, constants are much easier to

deal with.

> > I'm not sure whether I should try to come up with something more

> > precise but which will deviate from the traditional prescription. From

> > my point of view it is not worth it because these cmavo should be

> > phased out, but I will attempt it if there is a demand for it.

>

> I would like to go on record as both not disliking the cmavo in question

> and preferring that as many things be well defined as possible.

I'll see what I can do. But I doubt that the "fractional

quantifiers" on sets can at the same time be well

defined and in agreement with Lojban lore.

> Why do you dislike those other gadri so much, anyways?

Because:

a) They are redundant. Anything you can say with them can be

said by other means, usually in more simple terms.

b) They force you to make distinctions that you don't normally

want to make. We have pu/ca/ba for tenses, but we have the option

of using no tense when tense is unimportant or when tense makes

no sense. Similarly we need a non-content gadri for the occasions

when no gadri distinction is important or makes sense. The

proposed lo is such a gadri.

mu'o mi'e xorxes

'__

Do you Yahoo!?

SBC Yahoo! - Internet access at a great low price.

http://promo.yahoo.com/sbc/



Posted by Anonymous on Wed 19 of May, 2004 19:30 GMT

On Wed, May 19, 2004 at 11:26:24AM -0700, Jorge Llamb?as wrote:

>

> --- Robin Lee Powell wrote:

> > On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

> > > {le} and {la} become constants instead of being automatically

> > > quantified as {ro le} and {ro la}.

> >

> > What do you mean by 'constants'?

>

> Constants are direct references, not quantification over some set.

In English, for us little people?

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Wed 19 of May, 2004 19:53 GMT


> On Wed, May 19, 2004 at 11:26:24AM -0700, Jorge Llamb?as wrote:

> > --- Robin Lee Powell wrote:

> > > On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

> > > > {le} and {la} become constants instead of being automatically

> > > > quantified as {ro le} and {ro la}.

> > > What do you mean by 'constants'?

> > Constants are direct references, not quantification over some set.

> In English, for us little people?

Perhaps this can be of some help:

http://www.earlham.edu/~peters/courses/log/terms3.htm

mu'o mi'e xorxes

'__

Do you Yahoo!?

SBC Yahoo! - Internet access at a great low price.

http://promo.yahoo.com/sbc/



Posted by Anonymous on Wed 19 of May, 2004 20:29 GMT

On Wed, May 19, 2004 at 12:48:38PM -0700, Jorge Llamb?as wrote:

>

> --- Robin Lee Powell wrote:

> > On Wed, May 19, 2004 at 11:26:24AM -0700, Jorge Llamb?as wrote:

> > > --- Robin Lee Powell wrote:

> > > > On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

> > > > > {le} and {la} become constants instead of being automatically

> > > > > quantified as {ro le} and {ro la}.

> > > > What do you mean by 'constants'?

> > > Constants are direct references, not quantification over some set.

> > In English, for us little people?

>

> Perhaps this can be of some help:

> http://www.earlham.edu/~peters/courses/log/terms3.htm

Constants are short names or abbreviations for longer names; hence

"s" is a constant when it is used to abbreviate "Socrates".

So "lo mlatu", if it's a constant, should stand for some particular,

individual cat or cats, correct? This seems in contradiction with your

proposal:

The resulting expression refers generically to any individual or

group that satisfies the predicate.

I apoologise if I'm just being dumb.

-Robin



Posted by Anonymous on Wed 19 of May, 2004 21:18 GMT


> Constants are short names or abbreviations for longer names; hence

> "s" is a constant when it is used to abbreviate "Socrates".

>

> So "lo mlatu", if it's a constant, should stand for some particular,

> individual cat or cats, correct?

Or "Mr. Cat" himself, yes. But it is not a quantification over

the set of cats.

> This seems in contradiction with your

> proposal:

>

> The resulting expression refers generically to any individual or

> group that satisfies the predicate.

>

> I apoologise if I'm just being dumb.

{su'o lo mlatu} is a quantified expression over the set of cats.

{lo mlatu} is a constant term, which refers to the individual "cat(s)".

mi nelci lo mlatu

"I like cat(s)"

Not: "Given the set of all cats, there is at least one member x

of the set such that I like x". That's {mi nelci su'o mlatu}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

SBC Yahoo! - Internet access at a great low price.

http://promo.yahoo.com/sbc/


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Wed 19 of May, 2004 23:23 GMT posts: 14214

> Anonymous:

>

> --- Robin Lee Powell wrote:

> > Constants are short names or abbreviations for longer names; hence

> > "s" is a constant when it is used to abbreviate "Socrates".

> >

> > So "lo mlatu", if it's a constant, should stand for some particular,

> > individual cat or cats, correct?

>

> Or "Mr. Cat" himself, yes.

Those seem to be completely different kinds of things to me.

Please note that these aren't objections, I'm just trying to understand better.

-Robin


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Wed 19 of May, 2004 23:27 GMT posts: 14214

I'd like to see fewer examples of bare lo, and more of quantified lo please. Specificially, examples with just inner quantifiers and other examples with just outer quantifiers.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 00:22 GMT

Jorge "Llamb����������������������������������" wrote:

>--- Robin Lee Powell wrote:

>

>

>>On Wed, May 19, 2004 at 05:42:53AM -0700, Jorge Llamb?as wrote:

>>

>>

>>>{le} and {la} become constants instead of being automatically

>>>quantified as {ro le} and {ro la}.

>>>

>>>

>>What do you mean by 'constants'?

>>

>>

>

>Constants are direct references, not quantification over some set.

>

>With constants, you single out an individual (or group) and

>give a relationship in which it participates.

>

>With quantification you single out a set, and then you say

>how the members of the set are distributed with respect to

>a relationship.

>

Correct me if I'm wrong (and if I'm suffering from "when you have a

hammer everything looks like a nail" disease), but this sounds like an

intension/extension distinction. Your "constant" is referring to some

particular item(s) in extension, while your quantification talks about

"members of the set," i.e. the set they're in is what's important:

intension.

As I recall, gadri were rightly regarded as a big mess, and

intension/extension problems were one contributing factor to that. It

doesn't seem reasonable that what we all considered such a disaster

could be fixed by just a tiny change in default quantifiers and {lo} and

such.

>>>I'm not sure whether I should try to come up with something more

>>>precise but which will deviate from the traditional prescription. From

>>>my point of view it is not worth it because these cmavo should be

>>>phased out, but I will attempt it if there is a demand for it.

>>>

>>>

>>I would like to go on record as both not disliking the cmavo in question

>>and preferring that as many things be well defined as possible.

>>

Defined is good. This is Lojban. If we can't define it, we might as

well pack up and go home.

~mark



Posted by Anonymous on Thu 20 of May, 2004 01:12 GMT

On Wed, May 19, 2004 at 07:46:07PM -0400, Mark E. Shoulson wrote:

> Jorge "Llamb??????????????????????????????????????????????????????????????????????????????????????????????????????" wrote:

I think your mailer is confused, Mark. ---^

> As I recall, gadri were rightly regarded as a big mess, and

> intension/extension problems were one contributing factor to that. It

> doesn't seem reasonable that what we all considered such a disaster

> could be fixed by just a tiny change in default quantifiers and {lo}

> and such.

Just for the recond, this is *not* a tiny change. Removing all default

quantifiers is, in fact, a huge change.

It's just a change that seems to have a tiny *impact*. Different thing.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 01:12 GMT

Robin Lee Powell wrote:

>On Wed, May 19, 2004 at 07:46:07PM -0400, Mark E. Shoulson wrote:

>

>

>>Jorge "Llamb??????????????????????????????????????????????????????????????????????????????????????????????????????" wrote:

>>

>>

>

>I think your mailer is confused, Mark. ---^

>

Yeah, I wondered about that too. But it looked cool...

>>As I recall, gadri were rightly regarded as a big mess, and

>>intension/extension problems were one contributing factor to that. It

>>doesn't seem reasonable that what we all considered such a disaster

>>could be fixed by just a tiny change in default quantifiers and {lo}

>>and such.

>>

>>

>

>Just for the recond, this is *not* a tiny change. Removing all default

>quantifiers is, in fact, a huge change.

>

>It's just a change that seems to have a tiny *impact*. Different thing.

>

Mmm. Good point. Pondering the hugeness of the change is something I

need to do. But the question remains: if the impact is tiny, then in

what sense does it fix the big problem? That is, the prior text,

written under a "buggy" system, presumably needs some more serious

repair than a "tiny" impact.

I think the first difficulty I have is the same one you pointed out,

Robin: "some particular cat" and "Mr. Cat" are sufficiently different

critters that they deserve different gadri

~mark



Posted by Anonymous on Thu 20 of May, 2004 01:12 GMT


> Correct me if I'm wrong (and if I'm suffering from "when you have a

> hammer everything looks like a nail" disease), but this sounds like an

> intension/extension distinction. Your "constant" is referring to some

> particular item(s) in extension, while your quantification talks about

> "members of the set," i.e. the set they're in is what's important:

> intension.

I would have put it the other way around. Quantification ranges over

the extension of a set, whereas the constant I'm talking about is an

intensional object. Here is a fairly clear exposition on the

intensional/extensional distinction:

> As I recall, gadri were rightly regarded as a big mess, and

> intension/extension problems were one contributing factor to that. It

> doesn't seem reasonable that what we all considered such a disaster

> could be fixed by just a tiny change in default quantifiers and {lo} and

> such.

Well... If you want more details, there's another exposition of the

proposed system here:

The same proposal as written by And here:

(BTW, Robin, the links to that last page from other pages don't work,

probably because of something in its name, like the "--", do you think

it can be fixed somehow?)

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Thu 20 of May, 2004 01:29 GMT

On Wed, May 19, 2004 at 08:47:14PM -0400, Mark E. Shoulson wrote:

> Robin Lee Powell wrote:

> >>As I recall, gadri were rightly regarded as a big mess, and

> >>intension/extension problems were one contributing factor to that.

> >>It doesn't seem reasonable that what we all considered such a

> >>disaster could be fixed by just a tiny change in default quantifiers

> >>and {lo} and such.

> >>

> >>

> >

> >Just for the recond, this is *not* a tiny change. Removing all

> >default quantifiers is, in fact, a huge change.

> >

> >It's just a change that seems to have a tiny *impact*. Different

> >thing.

>

> Mmm. Good point. Pondering the hugeness of the change is something I

> need to do. But the question remains: if the impact is tiny, then in

> what sense does it fix the big problem? That is, the prior text,

> written under a "buggy" system, presumably needs some more serious

> repair than a "tiny" impact.

Show me:

1. Prior text that uses "lo broda" to mean something other than "broda

in general".

and/or

2. Prior text that uses inner quantifiers on lo.

AFAICT, those ore the only cases that change. There aren't many of

them (i.e. basically none).

> I think the first difficulty I have is the same one you pointed out,

> Robin: "some particular cat" and "Mr. Cat" are sufficiently different

> critters that they deserve different gadri

There are: lo pa mlatu versus lo mlatu. Unless I'm missing something.

I'm pretty sure I understand xorxes' proposal just fine, it's his

comments here that confused me. :-)

-Robin



Posted by Anonymous on Thu 20 of May, 2004 01:31 GMT


> Robin Lee Powell wrote:

> >Just for the recond, this is *not* a tiny change. Removing all default

> >quantifiers is, in fact, a huge change.

> >

> >It's just a change that seems to have a tiny *impact*. Different thing.

> >

> Mmm. Good point. Pondering the hugeness of the change is something I

> need to do.

The impact is almost exclusively for the better (i.e. it does not

invalidate previously valid text, but it does validate previously

suspect text.)

> But the question remains: if the impact is tiny, then in

> what sense does it fix the big problem? That is, the prior text,

> written under a "buggy" system, presumably needs some more serious

> repair than a "tiny" impact.

Basically, it solves the problem of intensional contexts by

providing another option besides quantification over the extension

of a set.

> I think the first difficulty I have is the same one you pointed out,

> Robin: "some particular cat" and "Mr. Cat" are sufficiently different

> critters that they deserve different gadri

Can we have examples please? Otherwise we get bogged down again in

meta-talk. How are they different? When do we use one or the other?

mi nelci su'o mlatu

There is some particular cat that I like.

mi nelci lo mlatu

I like cats.

ka'u lo mlatu cu kavbu lo smacu

(I know culturally:) Cats catch mice.

ju'ido'u za'a lo mlatu ca'o va kavbu lo smacu

Look! Cat(s) catching mouse/mice there!

{lo} is just an empty gadri, to be used when you don't need

to specify number, when you don't focus on the instances.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Thu 20 of May, 2004 01:33 GMT

On Wed, May 19, 2004 at 06:06:13PM -0700, Jorge Llamb?as wrote:

> The same proposal as written by And here:

>

>

> (BTW, Robin, the links to that last page from other pages don't work,

> probably because of something in its name, like the "--", do you think

> it can be fixed somehow?)

It is probably theoretically possible to fix it, but I don't care

enough, so I've renamed it. The problem is (and I don't know if this is

in the code or the browser) what was being looked for was not "--" but

"—".

http://www.lojban.org/tiki//XS%20gadri%20proposal:%20And's%20version

For links, XS gadri proposal: And's version should work. The Tiki

code seems to have already updated the pages that contained the old

name; go it.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 01:35 GMT

On Wed, May 19, 2004 at 06:30:52PM -0700, Jorge Llamb?as wrote:

> Can we have examples please? Otherwise we get bogged down again in

> meta-talk.

A-fucking-*MEN*.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 01:39 GMT

On Wed, May 19, 2004 at 06:30:52PM -0700, Jorge Llamb?as wrote:

> ju'ido'u za'a lo mlatu ca'o va kavbu lo smacu

> Look! Cat(s) catching mouse/mice there!

Why is that not using 'le'? Seems like if we're observing something, we

must have it in mind.

BTW, "COI DO" == "COI DO'U" in effect, but is one syllable shorter.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 02:01 GMT


> On Wed, May 19, 2004 at 06:30:52PM -0700, Jorge Llamb?as wrote:

> > ju'ido'u za'a lo mlatu ca'o va kavbu lo smacu

> > Look! Cat(s) catching mouse/mice there!

>

> Why is that not using 'le'? Seems like if we're observing something, we

> must have it in mind.

You can use {le} there, too. With {le}, what you're saying is:

"About that cat over there, and that mouse over there, I claim

that a kavbu relationship holds between them".

With {lo}, what you're saying is: "There's cat-catching-mouse

activity going on over there". If you don't particularly care to

say something about the particular cat and/or about the particular

mouse then you use {lo}.

The fact that you are seeing an instance of a cat does not mean

you are forced to say something about that instance. You can point

at the cat and mouse and say: "Didn't I tell you? Cats catch mice,

can you see it?"

> BTW, "COI DO" == "COI DO'U" in effect, but is one syllable shorter.

Yes, I just feel that using the second person pronoun in vocatives

is too abrupt. Natlang interference probably.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Thu 20 of May, 2004 04:45 GMT

On Wed, May 19, 2004 at 04:23:12PM -0700, [email protected] wrote:

> > > Constants are short names or abbreviations for longer names; hence

> > > "s" is a constant when it is used to abbreviate "Socrates".

> > >

> > > So "lo mlatu", if it's a constant, should stand for some particular,

> > > individual cat or cats, correct?

> >

> > Or "Mr. Cat" himself, yes.

>

> Those seem to be completely different kinds of things to me.

>

> Please note that these aren't objections, I'm just trying to understand better.

I have to agree. The rest of the gadri proposal makes perfect sense to me; but

as soon as Mr. Cat enters the picture, it sounds like something completely

different and alien.

I don't know when I'd want to talk about Mr. Cat, and I don't think I'd want to

accidentally be talking about him. I imagine that the concept will be similarly

difficult to grasp for most learners of the language.

It seems perfectly fine to define "lo" as an empty gadri. Why does Mr. Cat need

to be brought up at all?

--

Rob Speer



Posted by Anonymous on Thu 20 of May, 2004 04:45 GMT

On Thu, May 20, 2004 at 12:23:03AM -0400, Rob Speer wrote:

> On Wed, May 19, 2004 at 04:23:12PM -0700, [email protected]

> wrote:

> > > > Constants are short names or abbreviations for longer names;

> > > > hence "s" is a constant when it is used to abbreviate

> > > > "Socrates".

> > > >

> > > > So "lo mlatu", if it's a constant, should stand for some

> > > > particular, individual cat or cats, correct?

> > >

> > > Or "Mr. Cat" himself, yes.

> >

> > Those seem to be completely different kinds of things to me.

> >

> > Please note that these aren't objections, I'm just trying to

> > understand better.

>

> I have to agree. The rest of the gadri proposal makes perfect sense to

> me; but as soon as Mr. Cat enters the picture, it sounds like

> something completely different and alien.

Actually, from the perspective of the proposal I have no problem with

it, it's the 'constants' think that messes me up.

> I don't know when I'd want to talk about Mr. Cat, and I don't think

> I'd want to accidentally be talking about him. I imagine that the

> concept will be similarly difficult to grasp for most learners of the

> language.

Is there a difference between "ravens are tricksters" and "Raven is a

trickster"? Unless one is independetly worshiping Raven (in which case

it's "la raven" anyways), I don't see a difference.

> It seems perfectly fine to define "lo" as an empty gadri. Why does Mr.

> Cat need to be brought up at all?

See above. Also, this is apparently something that people feel the need

to say, although I'm not sure why.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 05:20 GMT

On Wed, May 19, 2004 at 09:35:41PM -0700, Robin Lee Powell wrote:

> > I don't know when I'd want to talk about Mr. Cat, and I don't think

> > I'd want to accidentally be talking about him. I imagine that the

> > concept will be similarly difficult to grasp for most learners of the

> > language.

>

> Is there a difference between "ravens are tricksters" and "Raven is a

> trickster"? Unless one is independetly worshiping Raven (in which case

> it's "la raven" anyways), I don't see a difference.

The words "Mr. Raven" simply don't explain anything to me. If "Mr. Raven"

means "an unspecified general bunch of ravens", then the definition

handles that case just fine without Mr. Raven needing to be mentioned.

"Mr. Raven" is an English translation of a concept that comes from some

other natlang, right? It's just not useful in talking about Lojban in

English, because your typical English reader will have no idea what it

means. Also, it takes a good concept and makes it sound silly.

Or is there some reason that it's essential that this unspecified general

bunch of ravens is considered as a single entity and given a name?

--

Rob Speer



Posted by Anonymous on Thu 20 of May, 2004 09:47 GMT

On Thu, May 20, 2004 at 01:02:49AM -0400, Rob Speer wrote:

> On Wed, May 19, 2004 at 09:35:41PM -0700, Robin Lee Powell wrote:

> > > I don't know when I'd want to talk about Mr. Cat, and I don't

> > > think I'd want to accidentally be talking about him. I imagine

> > > that the concept will be similarly difficult to grasp for most

> > > learners of the language.

> >

> > Is there a difference between "ravens are tricksters" and "Raven is

> > a trickster"? Unless one is independetly worshiping Raven (in which

> > case it's "la raven" anyways), I don't see a difference.

>

> The words "Mr. Raven" simply don't explain anything to me. If "Mr.

> Raven" means "an unspecified general bunch of ravens", then the

> definition handles that case just fine without Mr. Raven needing to be

> mentioned.

>

> "Mr. Raven" is an English translation of a concept that comes from

> some other natlang, right?

Wow, I'm sorry, it never occured to me that you simply hadn't

encountered the concept at all.

Raven is, depending on how you want to look at it, the animistic spirit

that embodies raven-ness, the Platonic ideal of raven-ness, or something

like that.

You might want to start with http://www.boisestate.edu/art/lmcneil/

> It's just not useful in talking about Lojban in English, because your

> typical English reader will have no idea what it means. Also, it takes

> a good concept and makes it sound silly.

Please note that Mr. Raven is not actually mentioned in the proposal.

> Or is there some reason that it's essential that this unspecified

> general bunch of ravens is considered as a single entity and given a

> name?

Many, many other languages do this to talk about classes of things in

general. I suspect, although I don't know this for fact, that in Native

American languages you don't say "All ravens are black", you say "Raven

is black". Certainly that's how stereotypical native americans talk

when speaking in english on TV.

-Robin



Posted by Anonymous on Thu 20 of May, 2004 13:05 GMT


> On Thu, May 20, 2004 at 12:23:03AM -0400, Rob Speer wrote:

> > The rest of the gadri proposal makes perfect sense to

> > me; but as soon as Mr. Cat enters the picture, it sounds like

> > something completely different and alien.

>

> Actually, from the perspective of the proposal I have no problem with

> it, it's the 'constants' think that messes me up.

A constant term is any term that is not a quantified term. If

it has no quantifier, it's a constant, that's all there is to it.

The word 'constant' is not mentioned in the definitions anyway.

The reason Mr. Broda comes up every time we discuss gadri is

that that's how JCB described generics in Loglan. Like it or

not, it's part of the language's traditional jargon. Again,

it is not mentioned in the definitions, so feel free to

ignore it.

Names are paradigmatic constant terms so thinking of "cats"

as "Mr. Cat" can clarify the logical structure of a proposition

for some people. If it doesn't work for you, just skip it.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer


Re: BPFK Section: gadri


[[user1|admin]] Posted by admin on Thu 20 of May, 2004 19:09 GMT posts: 208

This is actually by PC:

On {lo}

<< mi nitcu lo lanme i ko pirfi'i lo lanme seva'u mi

What I need is a sheep. Draw me a sheep.

mi djica lo lanme poi ba ze'u jmive

I want a sheep that will live a long time.>>

Even if {nitcu} and {djica} are in play yet again (are they?), it is hardly informative to use these examples, since they one of are wrong, prejudge a controversial issue, or need some explanation for folks who know how the corresponding words work in English (or just know about the paradoxes resulting from them as level with other contexts).

la

Name article. It converts a selbri, selecting its first argument, or a cmevla into a sumti. The resulting expression refers specifically to an individual or group that the speaker has in mind and which the speaker names with the selbri or cmevla. An outer quantifier can be used to quantify over members of the group. An inner quantifier can be used in the case of a selbri to indicate the cardinality of the group.

Can we fit a quantifier between {la} and the cmevla/selbri without getting the name wrong? How?

<

Typical article. It converts a selbri, selecting its first argument, into a sumti. The resulting expression refers to the typical individual or group that satisfies the predicate. An outer quantifier can be used to quantify over instances of the typical individual or group. An inner quantifier can be used to indicate the cardinality of the group.

le'e

Stereotypical article. It converts a selbri, selecting its first argument, into a sumti. The resulting expression refers to the stereotypical individual or group that is described by the predicate, from the point of view of the speaker. An outer quantifier can be used to quantify over instances of the stereotypical individual or group. An inner quantifier can be used to indicate the cardinality of the group.

>>

I am unsure just what to make of enumerating typical and stereotypical whatsises. The pattern seems to call for it, but in the cases where plurality plays a role, it seems to me that the typicality (etc.) is predicative not descriptive. That is you really want such things not just using the locution to talk in generalities about whatsises � whether or not they exist.

The treatment of quantifiers seems to me to be the sensible one and gets rid of 30 years of bootless disputes.

The examples in {lo}, combined with the dismissal of the {lo} = {so�u} equation (which probably should fade a bit) leaves {lo} ultimately unintelligible, if not contradictory.

pc



Posted by Anonymous on Thu 20 of May, 2004 19:48 GMT

(Repeating my answer due to accidental removal.)

pc:

> Can we fit a quantifier between {la} and the cmevla/selbri without getting

> the name wrong? How?

A quantifier between {la} and a cmevla is ungrammatical.

A quantifier between {la} and the selbri is grammatical.

The meaning of {la ci cribe} could be either "the three

that I call 'Bear'" or "the one that I call 'Three Bears'".

I'm not sure we really need to impose one of them.

> lo'e

> le'e

> >>

>

> I am unsure just what to make of enumerating typical and stereotypical

> whatsises. The pattern seems to call for it, but in the cases where plurality

> plays a role, it seems to me that the typicality (etc.) is predicative not

> descriptive. That is you really want such things not just using the locution

> to talk in generalities about whatsises � whether or not they exist.

I too am unsure what to make of them, but the grammar allows them.

I can leave the quantifiers out of lo'e/le'e if they are utterly

meaningless. Anyone else has an opinion on this?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Fri 21 of May, 2004 02:44 GMT

Robin Lee Powell scripsit:

> > I have to agree. The rest of the gadri proposal makes perfect sense to

> > me; but as soon as Mr. Cat enters the picture, it sounds like

> > something completely different and alien.

>

> Actually, from the perspective of the proposal I have no problem with

> it, it's the 'constants' think that messes me up.

The point is that since there's exactly one Mr. Cat, "all Mr. Cats" and

"some Mr. Cats" mean exactly the same thing, so you don't have to

worry about scope issues.

> Is there a difference between "ravens are tricksters" and "Raven is a

> trickster"? Unless one is independetly worshiping Raven (in which case

> it's "la raven" anyways), I don't see a difference.

Well, that's *one* meaning of English "Ravens are tricksters". It can

also mean that each raven is a trickster, or that some ravens are

tricksters, or the "thin abstraction" consisting of what ravens have

in common includes tricksterness.

--

Evolutionary psychology is the theory John Cowan

that men are nothing but horn-dogs, http://www.ccil.org/~cowan

and that women only want them for their money. http://www.reutershealth.com

--Susan McCarthy (adapted) [email protected]


Re: BPFK Section: gadri


[[user19|xod]] Posted by xod on Sat 22 of May, 2004 22:21 GMT posts: 143

doi xorxes .i .uofu'inai mi carmi zanru le selsni zo lo .i .oi ku'i le selsni zo loi zo'u

lu loi broda li'u

cu dunli

lu su'o2 broda li'u

.i xu la'e di'u cu jalge po'o le ro'efu'inaicai pensi

lo girzu ku {plurality}

.e lo gunma ku {mass}

.e lo selpau ku {substance}

.e loi dukse

mu'o mi'e xod



Posted by Anonymous on Sat 22 of May, 2004 22:43 GMT

On Sat, May 22, 2004 at 03:21:46PM -0700, [email protected] wrote:

> Re: BPFK Section: gadri

> doi xorxes .i .uofu'inai mi carmi zanru le selsni zo lo

Just for the record, that's not what carmi means. You want 'traji' or

'mutce'. Unless you are saying that your approval has a wavelength in

the visible spectrum?

> .e loi dukse

>

>

>

I wonder what happened to "mu'o mi'e xod", which I can see on the

discuss page.

-Robin



Posted by Anonymous on Sat 22 of May, 2004 23:20 GMT

> doi xorxes .i .uofu'inai mi carmi zanru le selsni zo lo .i .oi ku'i le selsni

> zo loi zo'u

coi xod mi do ckire lo nu pinka i mi spuda

> lu loi broda li'u

> cu dunli

> lu su'o2 broda li'u

sa'e lu loi broda li'u cu mutce simsa lu lo su'ore broda li'u

i lu su'ore broda cu brode li'u se smuni lo du'u su'o da poi

broda ku'o su'o de poi broda gi'e na du da zo'u da e de brode

i lu lo su'ore broda cu brode li'u se smuni lo du'u

lo su'oremei poi broda cu brode

ni'o ku'i le do pinka cu mapti i le me zo loi poi mi skicu ke'a

cu na mutce frica zo lo i ku'i lo bartu namcu cu ja'a frica stika

zo lo fa'u zo loi

> .i xu la'e di'u cu jalge po'o le ro'efu'inaicai pensi

> lo girzu ku {plurality}

> .e lo gunma ku {mass}

> .e lo selpau ku {substance}

> .e loi dukse

pe'i zo lo banzu lo nu tavla fi ro la'e di'u

i mu'a

lu la tenis ka'e te jivna lo re prenu lo re prenu li'u

lu ta culno lo bimu bidju li'u

lu lo tu'o djacu cu ta'e litki li'u

i lo nu zo lo basti zo loi cu na sidju lo nu jimpe

i do stidi ma

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer


Re: BPFK Section: gadri


Posted by xorxes on Sat 22 of May, 2004 23:57 GMT posts: 1912

> i lo nu zo lo basti zo loi cu na sidju lo nu jimpe

oi se'i mi pu skudji lu zo loi basti zo lo li'u

mu'o mi'e xorxes



Posted by Anonymous on Mon 24 of May, 2004 03:35 GMT

Jorge Llamb=EDas wrote:

>>doi xorxes .i .uofu'inai mi carmi zanru le selsni zo lo .i .oi ku'i le =

selsni

>>zo loi zo'u=20

>> =20

>>

>

>coi xod mi do ckire lo nu pinka i mi spuda

>=20

> =20

>

>>lu loi broda li'u=20

>>cu dunli=20

>>lu su'o2 broda li'u

>> =20

>>

>

>sa'e lu loi broda li'u cu mutce simsa lu lo su'ore broda li'u

>i lu su'ore broda cu brode li'u se smuni lo du'u su'o da poi

>broda ku'o su'o de poi broda gi'e na du da zo'u da e de brode

>i lu lo su'ore broda cu brode li'u se smuni lo du'u=20

>lo su'oremei poi broda cu brode

>

>ni'o ku'i le do pinka cu mapti i le me zo loi poi mi skicu ke'a=20

>cu na mutce frica zo lo i ku'i lo bartu namcu cu ja'a frica stika=20

>zo lo fa'u zo loi =20

> =20

>

je'eru'e .i ma mupli lo nu djica

lu 2 broda li'u

fa'u lu loi 2 broda li'u

fa'i lu lo 2 broda li'u

..i ma selkai loi broda .enai lo broda

>>.i xu la'e di'u cu jalge po'o le ro'efu'inaicai pensi=20

>>lo girzu ku {plurality}

>>.e lo gunma ku {mass}

>>.e lo selpau ku {substance}

>>.e loi dukse

>> =20

>>

>

>pe'i zo lo banzu lo nu tavla fi ro la'e di'u=20

>i mu'a=20

> lu la tenis ka'e te jivna lo re prenu lo re prenu li'u

> lu ta culno lo bimu bidju li'u=20

> =20

>

..i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne lo=20

si'o gunma

> lu lo tu'o djacu cu ta'e litki li'u

>i lo nu zo lo basti zo loi cu na sidju lo nu jimpe=20

> =20

>

..i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Mon 24 of May, 2004 17:18 GMT

xod:

> je'eru'e .i ma mupli lo nu djica

> lu 2 broda li'u

> fa'u lu loi 2 broda li'u

> fa'i lu lo 2 broda li'u

re prenu cu pleji lo rupnu be li pa do

i ja'ebo do cpacu lo rupnu be li re

i lo re prenu cu pleji lo rupnu be li pa do

i ja'ebo do cpacu lo rupnu be li pa

> .i ma selkai loi broda .enai lo broda

pe'i noda

....

> .i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne lo

> si'o gunma

do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e

pilno zo loi lo nu basna lo ka su'oremei

....

> .i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

mi senpi i pe'i ei zo tu'o se ciksi fi'o tcita zo tu'o

enai zo lo i mi na djica lo nu su'o prenu na zanru le

papri ki'u lo na mutce srana i ku'i mi ba pensi

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT

Jorge Llamb=EDas wrote:

>xod:

> =20

>

>>je'eru'e .i ma mupli lo nu djica

>>lu 2 broda li'u

>>fa'u lu loi 2 broda li'u

>>fa'i lu lo 2 broda li'u

>> =20

>>

>

>re prenu cu pleji lo rupnu be li pa do=20

>i ja'ebo do cpacu lo rupnu be li re

>i lo re prenu cu pleji lo rupnu be li pa do

>i ja'ebo do cpacu lo rupnu be li pa

> =20

>

..i'u je'ecai

>>.i ma selkai loi broda .enai lo broda

>> =20

>>

>

>pe'i noda

>

>...

> =20

>

>>.i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne lo

>>si'o gunma

>> =20

>>

>

>do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e=20

>pilno zo loi lo nu basna lo ka su'oremei

> =20

>

bi'unai lo selsni be zo loi cu nalsatci .i'enai .oisai

..i lo si'o gunma ku (noi nalsatci zi'e ne zo loi)

cu da'inai di'i mintu

lo si'o girzu ku (noi smuni satci zi'e ne .ei lu lo su'o broda li'u)

..i ji'a da'inai di'i mintu

lo si'o selmai ku (noi smuni satci zi'e ne .ei lu lo tu'o broda li'u=20

..onai lu tu'o broda li'u)

..i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

ge le du'u zo loi cu selbetri seldapma .iucu'i

gi tu'a zo tu'o

>...

> =20

>

>>.i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

>> =20

>>

>

>mi senpi i pe'i ei zo tu'o se ciksi fi'o tcita zo tu'o

>enai zo lo i mi na djica lo nu su'o prenu na zanru le

>papri ki'u lo na mutce srana i ku'i mi ba pensi

> =20

>

..ie mi zmanei

lo tcita be fi zo tu'o be'o

lo tcita be fi lu lo tu'o be'o

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT

Jorge Llamb=EDas wrote:

>do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e=20

>pilno zo loi lo nu basna lo ka su'oremei

> =20

>

la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u su'omei .i xy. jinvi=20

le du'u malgli

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT

xod:

> .i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

> ge le du'u zo loi cu selbetri seldapma .iucu'i

> gi tu'a zo tu'o

zo tu'o cmavo zo pa i mi ca'o finti lo velski be lo cmavo

be zo le joi zo la i ku'i vi'o do'u mi ka'e jmina lo pinka

be zo tu'o le papri cnita i zo loi zo'u mi na ka'e bilgygau

da lo nu na pilno i ro da jdice i mi finti lo velski poi

jbirai le tcaci velski lo se kakne be mi

mu'o mi'e xorxes

'__

Do you Yahoo!?

Yahoo! Domains – Claim yours for only $14.70/year

http://smallbusiness.promotions.yahoo.com/offer



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT


> la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u

> su'omei .i xy. jinvi le du'u malgli

iseki'ubo mi na ba stidi la'e di'u

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT

Jorge Llamb=EDas wrote:

>xod:

> =20

>

>>.i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

>>ge le du'u zo loi cu selbetri seldapma .iucu'i

>>gi tu'a zo tu'o

>> =20

>>

>

>zo tu'o cmavo zo pa i mi ca'o finti lo velski be lo cmavo=20

>be zo le joi zo la i ku'i vi'o do'u mi ka'e jmina lo pinka=20

>be zo tu'o le papri cnita i zo loi zo'u mi na ka'e bilgygau=20

>da lo nu na pilno i ro da jdice i mi finti lo velski poi=20

>jbirai le tcaci velski lo se kakne be mi=20

> =20

>

..iecai .i lo pu'u nalpli zo loi kei zo'u

norcatni je nalbilga je certu stidi ki'u lo du'u nalsatci

..i zo tu'o zo'u

do na galfi lo catni selsni gi'eku'i stidi le nu pilno tai lo gadri

..i ku'i la'e di'u .e la'e de'u cu selni'i .eise'anai la'o {excellent=20

solution} .i'ecai noi do ba'o stidi .e'edaisai

--

ni'o ta'o mi fliba troci lo nu fanta tu'a zoi {=3D20}

mu'o mi'e xod

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT

Jorge Llamb=EDas wrote:

>--- xod wrote:

>

> =20

>

>>la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u=20

>>su'omei .i xy. jinvi le du'u malgli

>> =20

>>

>

>iseki'ubo mi na ba stidi la'e di'u

>

>mu'o mi'e xorxes

> =20

>

..ienai mi jimpe tu'a lo ka ce'u su'omei ku goi sy. le do stidi vi la'o

{http://www.lojban.org/tiki//3DBPFK+Section%3A+gadri}

mu'o mi'e xod

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Mon 24 of May, 2004 22:47 GMT


> Jorge Llamb=EDas wrote:

> >--- xod wrote:

> >

> >>la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u=20

> >>su'omei .i xy. jinvi le du'u malgli

> >

> >iseki'ubo mi na ba stidi la'e di'u

>

> .ienai mi jimpe tu'a lo ka ce'u su'omei ku goi sy. le do stidi vi la'o

> {http://www.lojban.org/tiki//3DBPFK+Section%3A+gadri}

ku'i la'e di'u na mintu lo glico ka su'oremei i gy mapti

ge lu su'ore broda li'u gi lu lo su'ore broda li'u iseki'ubo

mi na tavla fi zoi gy plural gy

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 04:10 GMT

I'm going to go through and turn these into English. If this bothers

the authors, they are welcome to say so.

On Sun, May 23, 2004 at 11:23:57PM -0400, xod wrote:

> Jorge Llamb=EDas wrote:

>

> >>doi xorxes .i .uofu'inai mi carmi zanru le selsni zo lo .i .oi ku'i

> >>le selsni zo loi zo'u

xod:

Dear xorxes, I intensely approve of your meaning for "lo", but about "loi":

> >coi xod mi do ckire lo nu pinka i mi spuda

xorxes:

Dear xod, I appreciate your comment and I reply.

> >>lu loi broda li'u

> >>cu dunli

> >>lu su'o2 broda li'u

xod:

"loi broda" == "su'o 2 broda"

> >sa'e lu loi broda li'u cu mutce simsa lu lo su'ore broda li'u i lu

> >su'ore broda cu brode li'u se smuni lo du'u su'o da poi broda ku'o

> >su'o de poi broda gi'e na du da zo'u da e de brode i lu lo su'ore

> >broda cu brode li'u se smuni lo du'u lo su'oremei poi broda cu brode

xorxes:

Precisely speaking, "loi broda" is very similar to "lo su'o re broda".

"su'o re broda cu brode" means that not translating here su'o da poi

broda ku'o su'o do poi broda gi'e na du da zo'u da .e de broda

[[translation:%20there%20is%20at%20least%20one%20X,%20a%20broda,%20and%20at%20least%20one%20Y,%20a%0A%3Cbr%20/%3Ebroda%20that%20is%20not%20the%20same%20as%20X,%20where%20both%20are%20brode|translation: there is at least one X, a broda, and at least one Y, a

broda that is not the same as X, where both are brode]]. "lo su'o re

broda cu brode" means that lo su'oremei poi broda cu brode [[translation:%0A%3Cbr%20/%3EAn%20at-least-two-some%20which%20are%20broda%20are%20also%20brode|translation:

An at-least-two-some which are broda are also brode]].

> >ni'o ku'i le do pinka cu mapti i le me zo loi poi mi skicu ke'a cu na

> >mutce frica zo lo i ku'i lo bartu namcu cu ja'a frica stika zo lo

> >fa'u zo loi

xorxes still:

However, your comment fits. The "loi"-stuff which I describe is not

much different from "lo". However, the outer numbers [[i.e.%20outer%0A%3Cbr%20/%3Equantifiers|i.e. outer

quantifiers]] indeed differently adjust "lo" than "loi".

> je'eru'e .i ma mupli lo nu djica

> lu 2 broda li'u

> fa'u lu loi 2 broda li'u

> fa'i lu lo 2 broda li'u

xod:

Uhhh, OK. What is an example illustrating the event of desiring

I don't follow that, btw:

"2 broda"

and repsectively "loi 2 broda"

reciprocal of?? "lo 2 broda"

I'm%20assuming%20that%20both%20be jo'u or fa'u or similar. xod?" rel="">I'm assuming that both "fa'u" and "fa'i" were more-or-less intended to

be jo'u or fa'u or similar. xod?

> .i ma selkai loi broda .enai lo broda

What are the properties of "loi broda" that "lo broda" does not have?

rlpowell:

la'a zo loi cu po'o rinju le se smuni be zo lo .i va'i ro selkai be zo

loi cu vasru tu'a zo lo .iku'i ro selkai be zo lo cu ba'e na vasru tu'a

zo loi

> >>.i xu la'e di'u cu jalge po'o le ro'efu'inaicai pensi lo girzu ku

> >>{plurality}

> >>.e lo gunma ku {mass}

> >>.e lo selpau ku {substance}

> >>.e loi dukse

xod:

Is this a result only of intense thoughts about pluralities and masses

and substances and other things? Not sure about that last bit.

> >pe'i zo lo banzu lo nu tavla fi ro la'e di'u

> >i mu'a

> > lu la tenis ka'e te jivna lo re prenu lo re prenu li'u

> > lu ta culno lo bimu bidju li'u

> >

> >

xorxes:

I think "lo" suffices to talk about all of the above. For example:

"la tenis ka'e te jivna lo re prenu lo re prenu"

Tennis is a competition between two people and two people

"ta culno lo bimu bidju"

That is completely filled with half-beads.

> .i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne

> lo si'o gunma

xod:

Indeed, the idea of masses is part of the idea of pluralities, but what

is the fate of "loi", which is the adea of masses?

> > lu lo tu'o djacu cu ta'e litki li'u

> >i lo nu zo lo basti zo loi cu na sidju lo nu jimpe

xorxes:

more examples

"lo tu'o djacu cu ta'e litki"

I%20have%20*no*%20idea%20what%20to%20do%20with%20 nothing-water?" rel="">I have *no* idea what to do with "lo tu'o djacu". The

nothing-water?

"lo" substituting for "loi" doesn't help understanding.

> .i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

I approve of your use of "tu'o"; please suggest it to the BPFK.

Yes, please do; what are you talking about?

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:10 GMT

More translation.

On Mon, May 24, 2004 at 07:34:38AM -0700, Jorge Llamb?as wrote:

>

> xod:

> > je'eru'e .i ma mupli lo nu djica

> > lu 2 broda li'u

> > fa'u lu loi 2 broda li'u

> > fa'i lu lo 2 broda li'u

xod:

Uhhh, OK. What is an example illustrating the event of desiring

I don't follow that, btw:

"2 broda"

and repsectively "loi 2 broda"

reciprocal of?? "lo 2 broda"

I'm%20assuming%20that%20both%20be jo'u or fa'u or similar. xod?" rel="">I'm assuming that both "fa'u" and "fa'i" were more-or-less intended to

be jo'u or fa'u or similar. xod?

> re prenu cu pleji lo rupnu be li pa do

> i ja'ebo do cpacu lo rupnu be li re

> i lo re prenu cu pleji lo rupnu be li pa do

> i ja'ebo do cpacu lo rupnu be li pa

xorxes:

Two people pay one dollar to you. Therefore, you get two dollars. A

two-person group pays one dollar to you. Therefore, you get one dollar.

The former being "re prenu", the latter being "lo re prenu"

doi xorxes ne'i le do se stidi xu lu re prenu li'u cu dunli lu re lo

prenu li'u

> > .i ma selkai loi broda .enai lo broda

xod:

What are the properties of "loi broda" that "lo broda" does not have?

> pe'i noda

xorxes:

I think none at all.

> ...

> > .i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne

> > lo si'o gunma

xod:

Indeed, the idea of masses is part of the idea of pluralities, but what

is the fate of "loi", which is the adea of masses?

> do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e pilno

> zo loi lo nu basna lo ka su'oremei

xorxes:

What are you suggesting? I am satisfied with not using "loi", but I

suppose one could use "loi" to emphasize the property of being a

two-or-more-some.

> ...

> > .i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

xod:

I approve of your use of "tu'o"; please suggest it to the BPFK.

> mi senpi i pe'i ei zo tu'o se ciksi fi'o tcita zo tu'o enai zo lo i mi

> na djica lo nu su'o prenu na zanru le papri ki'u lo na mutce srana i

> ku'i mi ba pensi

I doubt it. I thing I am obligated to explain "tu'o" with the label of

"tu'o" and not "lo". explained." rel="">jatna: Unless you are expecting this usage to be a natural outcome of your "lo" proposal, in which case I expect it to be

explained. I don't want at least one person to disapprove of the page

because it was not relevant enough. But I will think on it.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:10 GMT

On Mon, May 24, 2004 at 02:02:21PM -0400, xod wrote:

snip

> >>.i ma selkai loi broda .enai lo broda

xod:

What are the properties of "loi broda" that "lo broda" does not have?

> >pe'i noda

xorxes:

I think none at all.

> >...

> >>.i'e lo si'o gunma cu pagbu lo si'o girzu .i ku'i ma dimna zo loi ne

> >>lo si'o gunma

xod:

Indeed, the idea of masses is part of the idea of pluralities, but what

is the fate of "loi", which is the adea of masses?

> >do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e

> >pilno zo loi lo nu basna lo ka su'oremei

xorxes:

What are you suggesting? I am satisfied with not using "loi", but I

suppose one could use "loi" to emphasize the property of being a

two-or-more-some.

> bi'unai lo selsni be zo loi cu nalsatci .i'enai .oisai

> .i lo si'o gunma ku (noi nalsatci zi'e ne zo loi)

> cu da'inai di'i mintu

> lo si'o girzu ku (noi smuni satci zi'e ne .ei lu lo su'o broda li'u)

> .i ji'a da'inai di'i mintu

> lo si'o selmai ku (noi smuni satci zi'e ne .ei lu lo tu'o broda li'u

> .onai lu tu'o broda li'u)

xod:

As I've said, the meaning of "loi" is not exact; this upsets me greatly

and I disapprove. The idea of a mass (which is inexact and is

associated with "lo") is in fact regularily identical with the idea of a

plurality (which has an exact meaning and must be "lo su'o broda"). In

addition, it is in fact regularily identical with the idea of a

substance (which has an exact meaning and must be "lo tu'o broda" or

"tu'o broda"). [[By%20which%20you%20mean%20a%20substance%20without%20intantiation%20in%20a%0A%3Cbr%20/%3Eparticular%20form?%20%20I%20guess?%20%20What's%20the%20difference%20between%20|By%20which%20you%20mean%20a%20substance%20without%20intantiation%20in%20a%0A%3Cbr%20/%3Eparticular%20form?%20%20I%20guess?%20%20What's%20the%20difference%20between%20]]and "no broda"?" rel="">By which you mean a substance without intantiation in a

particular form? I guess? What's the difference between "tu'o broda"

and "no broda"?

> .i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

> ge le du'u zo loi cu selbetri seldapma .iucu'i gi tu'a zo tu'o

xod still:

I conclude that I will not defend loi. Clearly you agree. Therefore

you must be honest with the BPFK that "loi"

is tragically cursed and about "tu'o".

> >...

> >>.i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

xod:

I approve of your use of "tu'o"; please suggest it to the BPFK.

> >mi senpi i pe'i ei zo tu'o se ciksi fi'o tcita zo tu'o enai zo lo i

> >mi na djica lo nu su'o prenu na zanru le papri ki'u lo na mutce srana

> >i ku'i mi ba pensi

xorxes:

I doubt it. I thing I am obligated to explain "tu'o" with the label of

"tu'o" and not "lo". [[jatna:%20Unless%20you%20are%20expecting%20this%20usage%20to%20be%0A%3Cbr%20/%3Ea%20natural%20outcome%20of%20your%20|%20Unless%20you%20are%20expecting%20this%20usage%20to%20be%0A%3Cbr%20/%3Ea%20natural%20outcome%20of%20your%20]]explained." rel="">jatna: Unless you are expecting this usage to be

a natural outcome of your "lo" proposal, in which case I expect it to be

explained. I don't want at least one person to disapprove of the page

because it was not relevant enough. But I will think on it.

> .ie mi zmanei

> lo tcita be fi zo tu'o be'o

> lo tcita be fi lu lo tu'o be'o

xod:

Yes, I prefer the label of "tu'o" to the lable of "lo tu'o".

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:10 GMT

On Mon, May 24, 2004 at 11:53:06AM -0700, Jorge Llamb?as wrote:

> xod:

> > .i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

> > ge le du'u zo loi cu selbetri seldapma .iucu'i gi tu'a zo tu'o

xod:

I conclude that I will not defend loi. Clearly you agree. Therefore

you must be honest with the BPFK that "loi" is tragically cursed and

about "tu'o".

> zo tu'o cmavo zo pa i mi ca'o finti lo velski be lo cmavo

> be zo le joi zo la i ku'i vi'o do'u mi ka'e jmina lo pinka

> be zo tu'o le papri cnita i zo loi zo'u mi na ka'e bilgygau

> da lo nu na pilno i ro da jdice i mi finti lo velski poi

> jbirai le tcaci velski lo se kakne be mi

xorxes:

"tu'o" is a cmavo of the class of "pa". I'm still making a description

of "le" and "la". However, I will comply; I am able to and comments

about tu'o to the bottom of the page. WRT "loi", I am not able to

obligate that it not be used. Everyone decides. I will invent a

description that is as near the customary one as I am able.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:19 GMT

On Mon, May 24, 2004 at 03:19:31PM -0400, xod wrote:

> Jorge Llamb=EDas wrote:

>

> >xod:

> >

> >

> >>.i ja'o mi na bandu zo loi .i za'a do tugni .isemu'ibo ko stace byfy

> >>ge le du'u zo loi cu selbetri seldapma .iucu'i

> >>gi tu'a zo tu'o

xod:

I conclude that I will not defend loi. Clearly you agree. Therefore

you must be honest with the BPFK that "loi" is tragically cursed and

about "tu'o".

> >zo tu'o cmavo zo pa i mi ca'o finti lo velski be lo cmavo be zo le

> >joi zo la i ku'i vi'o do'u mi ka'e jmina lo pinka be zo tu'o le papri

> >cnita i zo loi zo'u mi na ka'e bilgygau da lo nu na pilno i ro da

> >jdice i mi finti lo velski poi jbirai le tcaci velski lo se kakne be

> >mi

xorxes:

"tu'o" is a cmavo of the class of "pa". I'm still making a description

of "le" and "la". However, I will comply; I am able to and comments

about tu'o to the bottom of the page. WRT "loi", I am not able to

obligate that it not be used. Everyone decides. I will invent a

description that is as near the customary one as I am able.

> .iecai .i lo pu'u nalpli zo loi kei zo'u

> norcatni je nalbilga je certu stidi ki'u lo du'u nalsatci

xod:

Bravo! The process of not using "loi": neither the authority nor lack

of it, and not oblige and expertly suggest the fact on not being

precise.

> .i zo tu'o zo'u do na galfi lo catni selsni gi'eku'i stidi le nu pilno

> tai lo gadri

xod:

"tu'o": You do not modify the authoritative meaning, but you suggest a

usage with repsect to articles.

> .i ku'i la'e di'u .e la'e de'u cu selni'i .eise'anai la'o {excellent

> solution} .i'ecai noi do ba'o stidi .e'edaisai

xod:

But that and the other are implied (with obligation) by an "excellent

solution" which I am approvingly certain you will suggest, because you

are very good at this.

FWIW, I agree.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:19 GMT

On Mon, May 24, 2004 at 02:17:45PM -0400, xod wrote:

> Jorge Llamb=EDas wrote:

>

> >do stidi ma i mi se mansa lo nu na pilno zo loi i ku'i da'i ka'e

> >pilno zo loi lo nu basna lo ka su'oremei

xorxes:

What are you suggesting? I am satisfied with not using "loi", but I

suppose one could use "loi" to emphasize the property of being a

two-or-more-some.

> la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u su'omei .i xy.

> jinvi le du'u malgli

xod:

The Red Book I assume warns against using "loi" for the property of

two-some-ness. It suggests that that is malglico.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:19 GMT

On Mon, May 24, 2004 at 12:08:55PM -0700, Jorge Llamb?as wrote:

>

> --- xod wrote:

>

> > la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u su'omei .i xy.

> > jinvi le du'u malgli

xod:

The Red Book I assume warns against using "loi" for the property of

two-some-ness. It suggests that that is malglico.

> iseki'ubo mi na ba stidi la'e di'u

xorxes:

Therefore I will not suggest that.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:19 GMT

On Mon, May 24, 2004 at 03:25:11PM -0400, xod wrote:

> Jorge Llamb=EDas wrote:

> >--- xod wrote:

> >

> >>la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u

> >>su'omei .i xy. jinvi le du'u malgli

xod:

The Red Book I assume warns against using "loi" for the property of

two-some-ness. It suggests that that is malglico.

[[two-some-ness:%20That%20was%20an%20error,%20and%20should%20read%0A%3Cbr%20/%3E|two-some-ness: That was an error, and should read

"at-least-one-some-ness".]]

> >iseki'ubo mi na ba stidi la'e di'u

xorxes:

Therefore I will not suggest that.

> .ienai mi jimpe tu'a lo ka ce'u su'omei ku goi sy. le do stidi vi la'o

> {http://www.lojban.org/tiki//3DBPFK+Section%3A+gadri}

xod:

I disagree; I understand that the property of at-least-one-some-ness,

which we'll call S, is your suggestion in the BPFK gadri section.

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 04:19 GMT

On Mon, May 24, 2004 at 12:57:48PM -0700, Jorge Llamb?as wrote:

>

> --- xod wrote:

> > Jorge Llamb=EDas wrote:

> > >--- xod wrote:

> > >

> > >>la xunre cu kajde fi lo nu pilno zo loi lo ka ce'u=20 su'omei .i

> > >>xy. jinvi le du'u malgli

xod:

The Red Book I assume warns against using "loi" for the property of

at-least-one-some-ness. It suggests that that is malglico.

> > >iseki'ubo mi na ba stidi la'e di'u

xorxes:

Therefore I will not suggest that.

> > .ienai mi jimpe tu'a lo ka ce'u su'omei ku goi sy. le do stidi vi

> > la'o

> > {http://www.lojban.org/tiki//3DBPFK+Section%3A+gadri}

xod:

I disagree; I understand that the property of at-least-one-some-ness,

which we'll call S, is your suggestion in the BPFK gadri section.

> ku'i la'e di'u na mintu lo glico ka su'oremei i gy mapti ge lu su'ore

> broda li'u gi lu lo su'ore broda li'u iseki'ubo mi na tavla fi zoi gy

> plural gy

xorxes:

However, that's not the same as the English property of

at-least-two-some-ness. English fits both "su'o re broda" and

"lo su'o re broda". Therefore, I didn't say "plural".

-Robin

--

http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin.

"Many philosophical problems are caused by such things as the simple

inability to shut up." — David Stove, liberally paraphrased.

http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi



Posted by Anonymous on Tue 25 of May, 2004 06:47 GMT

On Mon, 24 May 2004, Robin Lee Powell wrote:

> I'm going to go through and turn these into English. If this bothers

> the authors, they are welcome to say so.

Quite good.

>

> > je'eru'e .i ma mupli lo nu djica

> > lu 2 broda li'u

> > fa'u lu loi 2 broda li'u

> > fa'i lu lo 2 broda li'u

>

> xod:

> Uhhh, OK. What is an example illustrating the event of desiring

> I don't follow that, btw:

> "2 broda"

> and repsectively "loi 2 broda"

> reciprocal of?? "lo 2 broda"

>

> I'm%20assuming%20that%20both%20> be jo'u or fa'u or similar. xod?" rel="">I'm assuming that both "fa'u" and "fa'i" were more-or-less intended to

> be jo'u or fa'u or similar. xod?

Whoops. fa'i should be fa'u.

> > > lu lo tu'o djacu cu ta'e litki li'u

> > >i lo nu zo lo basti zo loi cu na sidju lo nu jimpe

>

> xorxes:

> more examples

> "lo tu'o djacu cu ta'e litki"

> I%20have%20*no*%20idea%20what%20to%20do%20with%20> nothing-water?" rel="">I have *no* idea what to do with "lo tu'o djacu". The

> nothing-water?

> "lo" substituting for "loi" doesn't help understanding.

>

>

> > .i .i'e pilno zo tu'o noi ko .e'o stidi tecu'u byfy

>

> I approve of your use of "tu'o"; please suggest it to the BPFK.

>

> Yes, please do; what are you talking about?

tu'o doesn't mean zero, but is a number that means no number of

applicable. tu'o goes where a xo would deserve na'i. Substances are

uncountable; we can count arbitrary chunks (3 gallons, 8 grams) but the

chunks themselves don't obey expected numerical properties (one blob of

water, plus another, equals still only one blob.)

mu'o mi'e xod

--

"The Americans promised freedom and prosperity; what's this? Go up to

their headquarters, at one of those checkpoints where they point their

guns at you, and tell them that you hate them as much as Saddam, and see

what they do to you," said Mohammad Saleh, 39, a building contractor.

"The only difference is that Saddam would kill you in private, where the

Americans will kill you in public."



Posted by Anonymous on Tue 25 of May, 2004 06:47 GMT

On Mon, 24 May 2004, Robin Lee Powell wrote:

> On Mon, May 24, 2004 at 03:19:31PM -0400, xod wrote:

>

> > .iecai .i lo pu'u nalpli zo loi kei zo'u

> > norcatni je nalbilga je certu stidi ki'u lo du'u nalsatci

>

> xod:

> Bravo! The process of not using "loi": neither the authority nor lack

> of it, and not oblige and expertly suggest the fact on not being

> precise.

The phasing-out of 'loi' is an unofficial, unobligatory, expert suggestion

based on the fact that it's ambiguous.

> > .i ku'i la'e di'u .e la'e de'u cu selni'i .eise'anai la'o {excellent

> > solution} .i'ecai noi do ba'o stidi .e'edaisai

>

> xod:

> But that and the other are implied (with obligation) by an "excellent

> solution" which I am approvingly certain you will suggest, because you

> are very good at this.

It is his proposal of the Excellent Solution to the BF that was skillful.

(.oinai I try to be precise with the location of my .ui)

--

"The Americans promised freedom and prosperity; what's this? Go up to

their headquarters, at one of those checkpoints where they point their

guns at you, and tell them that you hate them as much as Saddam, and see

what they do to you," said Mohammad Saleh, 39, a building contractor.

"The only difference is that Saddam would kill you in private, where the

Americans will kill you in public."



Posted by Anonymous on Tue 25 of May, 2004 17:24 GMT

Let me stress that my comments were directed at the use of {nitcu lo} and {djica lo} in place of {nitcu tu'a lo} and {djica tu'a lo} as stock examples, not at other problems with the new {lo}. As a solution to the opaque context "problem" in Lojban, the new {lo} is both inadequate and ineffective. It is inadequate because it does not "solve" the "problem" for any type of sumti other than one introduced by {lo}, but the same problem occurs with every other type of compound sumti (LO + (bridi)) or names. It is ineffective because the problem persists: the inference from the new {lo bridi} in situ to {su'o da poi (bridi) go'i ... da...} works with the new {lo} as well as with the old — lacking the blocking action of {tu'a}.

It is also superfluous since Lojban does not have a problem with opaque contexts; {tu'a} and the abstract phrases it covers deal with these contexts in an exemplary fashion. The problem is with Lojbanists who refuse to use these devices, either because they have not taken the time to understand them or just persist in following their home language habit of not making the Lojban distinction — often even claiming that that habit rather than the Lojban (logical) format is the correct view of reality.

To be sure, we could ignore opaque contexts if we were willing to make some modifications to Lojban. But I think the modifications are much worse than just learning to use the language as she is given. We could, for example, insist that only existent objects culd be needed, wanted, dreamed about and so on. But this would limit our ability to plan and invent and countless other things we do — and want to — do. Unless, of course, we quantify not over existents but over non-existents (even impossibles) as well. But then we have to introduce the distinction between existents and non- explicitly, since everything (in the outer domain sense) is available for quantification, but not all are available for acting upon. Alternatively (or additionally, for that matter) we could deny that generalization was an inference that Lojban supported and take it that the logic it copied was a free logic (parallelling but slightly different from the view that "every" does not have existential

import). Along a different line, we might insist that {nitcu} and {djica} and an indefinite number of other predicates have (unmarked) opaque places, places which interfere with generalization (and Leibniz's law) and then just learn to recognize those places when the need arises (this is, of course, what natural languages do for the most part) and drop the need for explicit markers on ordinary places. This solution has regularly been rejected, even by people who act as though it were in place.

On the larger issue — which I did not mention except in passing — of the new definition of {lo} in general, I have said my say on that many times. Briefly, the definition is simply muddled since it tries to deal with a number of separate (and already fairly well separated in Lojban) concepts in a single device. The result is that the whole is at least incoherent and, I think I have shown, inconsistent. It does point to a couple of possible additions for Lojban: a way of talking about kinds separate from talking about the collective of its members or unspecified members or (stereo)typical members and a way of talking about stuffsubstancegoo. Note that, insofar as the new {lo} does either of these things, it fails in its other purposes, since neither the kind nor the kind-stuff does what an unspecified individual does. As for "generic" individuals, I have trouble seeing them as other than either the genus itself or a concrete individual without specifying which one, a

particular quantifier — with short scope, perhaps. The notion of Mr. Whatsis — at least the one that turns up most often — as a single object doing whatever any whatsis does really is contradictory and turns out to be nothing more than minimum-scoped quantifiers loosely disguised.

Ahah, another possibility for {lo}: it is just such a minimum-scoped quantifier and thus different from {su'o} which is, presumably, capable of being fronted. {lo} would then not be capable of being fronted and would — by itself — block generalization. Further, it would not be anaphorizable, since that would illegitimately extend its scope. (Actually, it would be anaphorizable with pronouns that repeat the phrase but not the referent of the phrase — as happens in English, for example). That might be a useful change to make, but I suspect that even it would go against much previous usage.

The possibility that {lo} really does refer to the genus — as opposed to the collective or the set or the members taken separately — might also work, but it would require a rewriting of the semantics of virtually every place since most places have been taken to deal with members of the genus, not the genus itself. And then the rewrites have to be adjusted to account for sumti not introduced by {lo} (or all the other gadri have to be redefined as well).

Frankly, I think the best solution to all the problems that the new {lo} is meant to solve is just learning to deal with Lojban itself, not Lojanized English or Spanish or whatever.

pc



Posted by Anonymous on Tue 25 of May, 2004 17:24 GMT

John E Clifford wrote:

>Let me stress that my comments were directed at the use of {nitcu lo} and {djica lo} in place of {nitcu tu'a lo} and {djica tu'a lo} as stock examples, not at other problems with the new {lo}. As a solution to the opaque context "problem" in Lojban, the new {lo} is both inadequate and ineffective.

>

>

It was my understanding that the new lo is opaque, not su'oda,

intentional, Mister, and suitable for "needed boxes" and off-white

unicorns. If it is not, or if those concepts are not identical

(rendering my understanding into a confused mess) then I and others will

need concrete use-cases showing the errors I've made, and the actual

difficulties this presents. (In fact, the response will have to be even

clearer than this very paragraph of mine, which unfortunately is above

the heads of many of the BF commissioners, who have not attended the

jboske seminar series; thus the need for concrete example sentences.)

mu'o mi'e xod

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Tue 25 of May, 2004 18:36 GMT


> xorxes:

> I think "lo" suffices to talk about all of the above. For example:

> "la tenis ka'e te jivna lo re prenu lo re prenu"

> Tennis is a competition between two people and two people

Or "Tennis can be played two against two."

> "ta culno lo bimu bidju"

> That is completely filled with half-beads.

{bimu} = 85, but it works with {pimu} as well.

Are {bi} and {pi} too similar to be both in PA?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 18:36 GMT


> doi xorxes ne'i le do se stidi xu lu re prenu li'u cu dunli lu re lo

> prenu li'u

go'i

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 18:36 GMT

These usages seem entirely non-problematic and, except possibly for the question of where the quantifiers go, don't obviously require the new {lo} (I suppose the scope rules are a bit vague here, but that seems to be mainly because no one has bothered to work them up)

pc

Jorge Llambías wrote:


> xorxes:

> I think "lo" suffices to talk about all of the above. For example:

> "la tenis ka'e te jivna lo re prenu lo re prenu"

> Tennis is a competition between two people and two people

Or "Tennis can be played two against two."

> "ta culno lo bimu bidju"

> That is completely filled with half-beads.

{bimu} = 85, but it works with {pimu} as well.

Are {bi} and {pi} too similar to be both in PA?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 19:11 GMT

(Many of pc's points refer to quantified terms. I agree with most of

what he says, but they don't concern the proposed lo, which is not a

quantified term.) I respond to some of the other points.

pc:

> As a solution to the opaque context

> "problem" in Lojban, the new {lo} is both inadequate and ineffective. It is

> inadequate because it does not "solve" the "problem" for any type of sumti

> other than one introduced by {lo}, but the same problem occurs with every

> other type of compound sumti (LO + (bridi)) or names.

Could you explain what the problem is with:

mi nitcu le va tanxe

I need that box.

I don't see what could be the difference between:

"I need: (For that box: I have it)" and

"For that box: (I need: I have it)". Is there a

difference?

...

> Alternatively (or additionally,

> for that matter) we could deny that generalization was an inference that

> Lojban supported and take it that the logic it copied was a free logic

> (parallelling but slightly different from the view that "every" does not have

> existential

> import).

Presumably any logic can be expressed in Lojban as well as in English

or any other language. It seems that generalization is something that

we can do or not do irrespective of the language in which we express

it.

...

> As for "generic" individuals, I have trouble seeing them as other than

> either the genus itself or a concrete individual without specifying which

> one, a

> particular quantifier — with short scope, perhaps. The notion of Mr.

> Whatsis — at least the one that turns up most often — as a single object

> doing whatever any whatsis does really is contradictory and turns out to be

> nothing more than minimum-scoped quantifiers loosely disguised.

Could you please point out the contradiction, ideally with an example?

...

> Frankly, I think the best solution to all the problems that the new {lo} is

> meant to solve is just learning to deal with Lojban itself, not Lojanized

> English or Spanish or whatever.

If you could write the proposed lo-examples in what you consider

more correct Lojban, that would help to highlight where the differences

are.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 19:59 GMT

By popular request, here are some examples of the non-universal

(prescribed) usage of {lo}. I will probably find more if I continue to

look.

do citme'a mi lo nanca be li xa

You're 6 years younger than me.

la stace pu citka lo cirla

=46rank ate some cheese. (or one or more pieces of cheese).

i abu ca'o menli jdice to sekai le xagrai selka'e pe va'o le nu le glar=

e djedi cu rinka le nu abu lifri le nu sipydji je bebna toi le du'u xuk=

au le nu pluka fa le nu zbasu lo xrula linsi cu se vamji le raktu poi n=

u sa'irbi'o gi'e crepu loi xrula icabo suksa fa le nu lo blabi ractu po=

i xunblabi se kanla cu bajra zo'a a bu

(lo blabi ractu =3D a (1) white rabbit)

--=20

Arnt Richard Johansen http://arj.nvg.org=

/

Jeg er nok verdens sydligste sengev=E6ter. Forutsatt at ingen p=E5 base=

n p=E5

Sydpolen driver med slikt, da. --Erling Kagge: Alene til Sydpole=

n



Posted by Anonymous on Tue 25 of May, 2004 19:59 GMT

Arnt Richard Johansen wrote:

>By popular request, here are some examples of the non-universal

>(prescribed) usage of {lo}. I will probably find more if I continue to

>look.

>

>

>

I find these in keeping with the proposed usage of lo, which I

understand to be similar to the English "any".

>do citme'a mi lo nanca be li xa

>You're 6 years younger than me.

>

>

A general six years; not a particular six years.

>la stace pu citka lo cirla

>=46rank ate some cheese. (or one or more pieces of cheese).

>

>

That's right. Not a particular chunk.

>i abu ca'o menli jdice to sekai le xagrai selka'e pe va'o le nu le glar=

>e djedi cu rinka le nu abu lifri le nu sipydji je bebna toi le du'u xuk=

>au le nu pluka fa le nu zbasu lo xrula linsi cu se vamji le raktu poi n=

>u sa'irbi'o gi'e crepu loi xrula icabo suksa fa le nu lo blabi ractu po=

>i xunblabi se kanla cu bajra zo'a a bu

>(lo blabi ractu =3D a (1) white rabbit)

>

>--=20

>Arnt Richard Johansen http://arj.nvg.org=

>/

>Jeg er nok verdens sydligste sengev=E6ter. Forutsatt at ingen p=E5 base=

>n p=E5

>Sydpolen driver med slikt, da. --Erling Kagge: Alene til Sydpole=

>n

>

>

>

>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Tue 25 of May, 2004 20:03 GMT posts: 14214

BTW, xod says that "lo broda poi brode" == "le broda" (with loss of information, of course).

Is that always true, or is it specific to your scheme? Regardless, it seems to be true and I'd appreciate it if you mentioned it somewhere.

In case it isn't obvious, conversion formula give me

W

Wmake me very happy.

-Robin



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT

[email protected] wrote:

>Re: BPFK Section: gadri

>BTW, xod says that "lo broda poi brode" == "le broda" (with loss of information, of course).

>

>Is that always true, or is it specific to your scheme? Regardless, it seems to be true and I'd appreciate it if you mentioned it somewhere.

>

>In case it isn't obvious, conversion formula give me

W

Wmake me very happy. >

>-Robin

>

>

>

I've found a dyad of ideas which approximate *Intension* and

  • Extension*, even if the correlation is not perfect. Intension is

similar to *selkaicfa*; la'e-like, beginning with a set of qualities,

one discusses the items, if any, that qualify. Extension is similar to

  • kaicfa*; lu'e-like, beginning with an object in mind, one discusses its

qualities in order to describe it. (When I use "any" in English, I am

experiencing selkaicfa; I have a requirement in mind, and I am referring

to whatever items fit the bill. However, I also can use "any" to express

a lack of preference among the members of a set, all of whose members

might be known specifically to me: "Hit this button with any of your

fingers.")

Under Jorge's proposal before the BF, lo = selkaicfa, le = kaicfa. With

selkaicfa, the speaker has given an explicit description of the required

qualities, and the sumti refers to anything that matches. In kaicfa, the

speaker has given only a partial list of the qualities, in order to aid

the listener in understanding.

So lo poi isn't le.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT


> Re: BPFK Section: gadri

> BTW, xod says that "lo broda poi brode" == "le broda" (with loss of

> information, of course).

>

> Is that always true, or is it specific to your scheme? Regardless, it seems

> to be true and I'd appreciate it if you mentioned it somewhere.

{le broda} can refer to anything at all that you have in mind, so

I don't understand exactly what you mean. In general it is not equivalent

to use one expression or the other. With {le} there is a certain thing

or things that you want to talk about, you have them identified, so you

will refer to them as {le ....}. All you need to do is pick a predicate

that will be helpful to your audience to identify what thing or things

you are making a claim about.

With {lo} you won't be making a claim about any specific thing or

things, you will only talk about the things that satisfy a certain

predicate, whether or not there are any of them around.

With {lo broda poi brode} you claim something about brodas

that brode. For example:

lo gerku poi cmoni cu na batci

A dog that barks won't bite.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT


> By popular request, here are some examples of the non-universal

> (prescribed) usage of {lo}. I will probably find more if I continue to

> look.

Are these supposed to be problematic for the proposed {lo}?

They are not.

> do citme'a mi lo nanca be li xa

> You're 6 years younger than me.

This one makes more sense with {lo} refering generically to

a period of six years rather than quantifying over periods

od six years: "there is at least one period of six years

such that you're younger than my by that period"? Not wrong,

but weird. The natural thing to say is that the difference

in our ages is the single entity "6 years".

> la stace pu citka lo cirla

> =46rank ate some cheese. (or one or more pieces of cheese).

That one works with both {lo cirla} or {su'o cirla}.

> i abu ca'o menli jdice to sekai le xagrai selka'e pe va'o le nu le glar=

> e djedi cu rinka le nu abu lifri le nu sipydji je bebna toi le du'u xuk=

> au le nu pluka fa le nu zbasu lo xrula linsi cu se vamji le raktu poi n=

> u sa'irbi'o gi'e crepu loi xrula icabo suksa fa le nu lo blabi ractu po=

> i xunblabi se kanla cu bajra zo'a a bu

> (lo blabi ractu =3D a (1) white rabbit)

(I would use {lo nu} for most of those {le nu} now.)

Yes, you can use {lo blabi ractu} when there is only one white

rabbit around.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT

On Tue, May 25, 2004 at 09:20:16PM +0200, Arnt Richard Johansen wrote:

> By popular request, here are some examples of the non-universal

> (prescribed) usage of {lo}. I will probably find more if I continue to

> look.

For those of you who weren't on IRC at the time, Arnt believes that

xorxes' proposal will invalidate much past usage of lo.

We asked him for examples; these are they.

All my interpretations are based on my understanding of xorxes'

proposal, which I admit is limited.

> do citme'a mi lo nanca be li xa

> You're 6 years younger than me.

You are less than me in any length-6-years time period.

Not seeing a problem.

> la stace pu citka lo cirla

> Frank ate some cheese. (or one or more pieces of cheese).

Honest ate any cheese.

Not seeing a problem; if you care which cheese he ate, use le.

> i abu ca'o menli jdice to sekai le xagrai selka'e pe va'o le nu le

> glare djedi cu rinka le nu abu lifri le nu sipydji je bebna toi le

> du'u xukau le nu pluka fa le nu zbasu lo xrula linsi cu se vamji le

> raktu poi nu sa'irbi'o gi'e crepu loi xrula icabo suksa fa le nu lo

> blabi ractu poi xunblabi se kanla cu bajra zo'a a bu

> (lo blabi ractu = a (1) white rabbit)

Where did "a (1)" come from? It has *never* been the case in Lojban

that "lo blabi ractu" means "one white rabbit".

-Robin



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT

On Tue, May 25, 2004 at 02:55:21PM -0700, Jorge Llamb?as wrote:

> > (lo blabi ractu =3D a (1) white rabbit)

>

> Yes, you can use {lo blabi ractu} when there is only one white rabbit

> around.

Please expand on this. In particular, if there is only one white rabbit

around and you say "lo blabi ractu", are you referring to said local

white rabbit in any useful way?

-Robin



Posted by Anonymous on Tue 25 of May, 2004 23:54 GMT


> On Tue, May 25, 2004 at 02:55:21PM -0700, Jorge Llamb?as wrote:

> > Yes, you can use {lo blabi ractu} when there is only one white rabbit

> > around.

>

> Please expand on this. In particular, if there is only one white rabbit

> around and you say "lo blabi ractu", are you referring to said local

> white rabbit in any useful way?

You're not referring to it specifically, no, but the fact that there

is a white rabbit present does not preclude that you speak about

white rabbits either. But this is independent of my proposal.

If you say {su'o blabi ractu} you are not referring to any local

rabbit either, you are quantifying over the set of all rabbits.

This has to do with specificity {le} vs. {su'o}/{lo}, not with

{su'o} vs. {lo}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

pc adds caps to indicate where replies fit.

A: Interesting. I wasn't aware of saying much of anything about quantified expressions — escept that one can be inferred from any unblocked sumti

B: There is very little the matter with {mi nitcu leva tanxe} aside from its taking the basket as an event. It does not create generalization problems because it has guaranteed — by explicit deixis — that the object referred to exists in this world. This clearly a very special case and is handled as such. I suspect that at a deep grammatical level it is handled — unlike most cases — by insertion from an external sumti.

C: My point is exactly that if we want to prevent the problems when we elect to do generalization we have to forbid generalization if we are not going to mark places where it does not apply, {nitcu2} for example.

D: Rabbit A is eating grass, Rabbit B is not, so Mr. Rabbit both is and is not eating grass. This is a flat contradiction and therefore, since contradictory objects cannot exist, Mr. Rabbit — as proposed — does not exist. And so, assuming that we have not decided to use the outer domain, Mr. Rabbit is not the referent of any {lo ractu} expression — except perhaps one in an intentional context.

E: already done God knows how often over the last few decades. Most of them are indistinguishable from cases of new {lo}, which is meant to appear conservative. The opaque cases — as with {nitcu} — are the most obvious differences. Other would probably be negative examples, where {lo} is used in place of some other more accurate gadri: {loi}, for example, or the possibly needed genus or goo forms.

pc

A:

(Many of pc's points refer to quantified terms. I agree with most of

what he says, but they don't concern the proposed lo, which is not a

quantified term.) I respond to some of the other points.

pc:

> As a solution to the opaque context

> "problem" in Lojban, the new {lo} is both inadequate and ineffective. It is

> inadequate because it does not "solve" the "problem" for any type of sumti

> other than one introduced by {lo}, but the same problem occurs with every

> other type of compound sumti (LO + (bridi)) or names.

B:

Could you explain what the problem is with:

mi nitcu le va tanxe

I need that box.

I don't see what could be the difference between:

"I need: (For that box: I have it)" and

"For that box: (I need: I have it)". Is there a

difference?

...

> Alternatively (or additionally,

> for that matter) we could deny that generalization was an inference that

> Lojban supported and take it that the logic it copied was a free logic

> (parallelling but slightly different from the view that "every" does not have

> existential

> import).

C:Presumably any logic can be expressed in Lojban as well as in English

or any other language. It seems that generalization is something that

we can do or not do irrespective of the language in which we express

it.

...

> As for "generic" individuals, I have trouble seeing them as other than

> either the genus itself or a concrete individual without specifying which

> one, a

> particular quantifier — with short scope, perhaps. The notion of Mr.

> Whatsis — at least the one that turns up most often — as a single object

> doing whatever any whatsis does really is contradictory and turns out to be

> nothing more than minimum-scoped quantifiers loosely disguised.

D:Could you please point out the contradiction, ideally with an example?

...

> Frankly, I think the best solution to all the problems that the new {lo} is

> meant to solve is just learning to deal with Lojban itself, not Lojanized

> English or Spanish or whatever.

E:If you could write the proposed lo-examples in what you consider

more correct Lojban, that would help to highlight where the differences

are.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

Careful. English "any" fluctuates (in fairly contoleed ways — but you need to supply more context) between universal and particlar. I take it this means particular "some unspecifed one(s)."

The other cases are merely conservative, that is not different from "old {lo}" though they may differ from some intermediate cases which arose once folks started mucking about with {lo} (this time or earlier).

pc

xod wrote:

I find these in keeping with the proposed usage of lo, which I

understand to be similar to the English "any".

>do citme'a mi lo nanca be li xa

>You're 6 years younger than me.

>

>

A general six years; not a particular six years.

>la stace pu citka lo cirla

>=46rank ate some cheese. (or one or more pieces of cheese).

>

>

That's right. Not a particular chunk.



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

I can't find xod's claim (but I know that some things never get to me — and some of mine don't get to some other people). It is however pretty certainly false for almost any reading of {lo}: {lo broda poi brode} is only marginally more specific than {lo broda} — and maybe not at all if all or even most broda are also brode: Any old broda poi brode is not a specified one except accidentally.

pc

[email protected] wrote:

Re: BPFK Section: gadri

BTW, xod says that "lo broda poi brode" == "le broda" (with loss of information, of course).

Is that always true, or is it specific to your scheme? Regardless, it seems to be true and I'd appreciate it if you mentioned it somewhere.

In case it isn't obvious, conversion formula give me

W

Wmake me very happy.

-Robin



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

pc:

> B: There is very little the matter with {mi nitcu leva tanxe} aside from its

> taking the basket as an event. It does not create generalization problems

> because it has guaranteed — by explicit deixis — that the object referred

> to exists in this world. This clearly a very special case and is handled as

> such. I suspect that at a deep grammatical level it is handled — unlike

> most cases — by insertion from an external sumti.

You had said that the new {lo} is "inadequate because it does not "solve"

the "problem" for any type of sumti other than one introduced by {lo},

but the same problem occurs with every other type of compound sumti

(LO + (bridi)) or names." What would be an example where {nitcu le} or

{nitcu la} are problematic, then?

> D: Rabbit A is eating grass, Rabbit B is not, so Mr. Rabbit both is and is

> not eating grass. This is a flat contradiction and therefore, since

> contradictory objects cannot exist, Mr. Rabbit — as proposed — does not

> exist. And so, assuming that we have not decided to use the outer domain,

> Mr. Rabbit is not the referent of any {lo ractu} expression — except

> perhaps one in an intentional context.

Mr Rabbit is eating grass (here), but he is not eating grass (over there).

That's not contradictory. John talked to Mary (yesterday), but he did not

talk to Mary (the day before). That's the same type of non-contradiction.

That John did and did not talk to Mary does not mean he cannot exist. All

it means is that to understand what a sentence means you need context.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

I don't quite see how this distinction fits the extension-intension distinction and even less what that distinction has to do with the {lo}-{le} contrast. Both {lo broda} and {le broda} pick out things with reference to the (probably very vague) property of being a broda. In one case, the speaker is apparently indifferent to or ignorant of which broda satisfies the rest of the sentence. In the other case, the speaker knows and wishes to let the hearer know which one it is and uses the predication to aid that task. Both are extensional (refer to things, not concepts). The distinction given here does seem to bear on the difference between {le} and {lo}, though the connection bertween these and {la'e} and {lu'e} is forced at best.

pc

xod wrote:

[email protected] wrote:

>Re: BPFK Section: gadri

>BTW, xod says that "lo broda poi brode" == "le broda" (with loss of information, of course).

>

>Is that always true, or is it specific to your scheme? Regardless, it seems to be true and I'd appreciate it if you mentioned it somewhere.

>

>In case it isn't obvious, conversion formula give me

W

Wmake me very happy. >

>-Robin

>

>

>

I've found a dyad of ideas which approximate *Intension* and

  • Extension*, even if the correlation is not perfect. Intension is

similar to *selkaicfa*; la'e-like, beginning with a set of qualities,

one discusses the items, if any, that qualify. Extension is similar to

  • kaicfa*; lu'e-like, beginning with an object in mind, one discusses its

qualities in order to describe it. (When I use "any" in English, I am

experiencing selkaicfa; I have a requirement in mind, and I am referring

to whatever items fit the bill. However, I also can use "any" to express

a lack of preference among the members of a set, all of whose members

might be known specifically to me: "Hit this button with any of your

fingers.")

Under Jorge's proposal before the BF, lo = selkaicfa, le = kaicfa. With

selkaicfa, the speaker has given an explicit description of the required

qualities, and the sumti refers to anything that matches. In kaicfa, the

speaker has given only a partial list of the qualities, in order to aid

the listener in understanding.

So lo poi isn't le.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."

't



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

There is of course a particular period of six years that you are younger than I but that is probably not what is meant here and so anyold period will do. I agree it would be nicer if we had a unit concept, but the way that "younger by six years " has developed makes that difficult ({lo nanca xamei} is weird in other ways, as is {lo xavnanca}).

pc

Jorge Llambías wrote:


> By popular request, here are some examples of the non-universal

> (prescribed) usage of {lo}. I will probably find more if I continue to

> look.

Are these supposed to be problematic for the proposed {lo}?

They are not.

> do citme'a mi lo nanca be li xa

> You're 6 years younger than me.

This one makes more sense with {lo} refering generically to

a period of six years rather than quantifying over periods

od six years: "there is at least one period of six years

such that you're younger than my by that period"? Not wrong,

but weird. The natural thing to say is that the difference

in our ages is the single entity "6 years".



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

E: {lo nitcu la meripapnz} since Mary Poppins does not exist. {nitcu le xamoi archon} which archon also does not exist for all that it is specific in my need ("controls the sphere of Mars") Most anything else with references to nonexistents will do — unless we have made some unmentioned cirumvention.

F: Sorry, but Mr. Rabbit (i.e., rabbit b) is not eating grass here and now just as he is eating it here and now (as rabbit a).

Jorge Llambías wrote:

pc:

> B: There is very little the matter with {mi nitcu leva tanxe} aside from its

> taking the basket as an event. It does not create generalization problems

> because it has guaranteed — by explicit deixis — that the object referred

> to exists in this world. This clearly a very special case and is handled as

> such. I suspect that at a deep grammatical level it is handled — unlike

> most cases — by insertion from an external sumti.

E:You had said that the new {lo} is "inadequate because it does not "solve"

the "problem" for any type of sumti other than one introduced by {lo},

but the same problem occurs with every other type of compound sumti

(LO + (bridi)) or names." What would be an example where {nitcu le} or

{nitcu la} are problematic, then?

> D: Rabbit A is eating grass, Rabbit B is not, so Mr. Rabbit both is and is

> not eating grass. This is a flat contradiction and therefore, since

> contradictory objects cannot exist, Mr. Rabbit — as proposed — does not

> exist. And so, assuming that we have not decided to use the outer domain,

> Mr. Rabbit is not the referent of any {lo ractu} expression — except

> perhaps one in an intentional context.

F:Mr Rabbit is eating grass (here), but he is not eating grass (over there).

That's not contradictory. John talked to Mary (yesterday), but he did not

talk to Mary (the day before). That's the same type of non-contradiction.

That John did and did not talk to Mary does not mean he cannot exist. All

it means is that to understand what a sentence means you need context.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 02:59 GMT

On Tue, May 25, 2004 at 05:05:13PM -0700, John E Clifford wrote:

> I can't find xod's claim

It was never posted here.

> (but I know that some things never get to me — and some of mine don't

> get to some other people).

If you have specific examples, please let me know.

-Robin



Posted by Anonymous on Wed 26 of May, 2004 03:00 GMT

On Tue, May 25, 2004 at 10:42:35AM -0700, Jorge Llamb?as wrote:

> --- Robin Lee Powell wrote:

> > "ta culno lo bimu bidju"

> > That is completely filled with half-beads.

>

> {bimu} = 85, but it works with {pimu} as well.

  • Whoops*.

> Are {bi} and {pi} too similar to be both in PA?

Oh, believe me, if we start down the path of trying to disentangle

easily mix-uppable Lojban words, we'll never get done. Let's not start.

-Robin



Posted by Anonymous on Wed 26 of May, 2004 03:00 GMT

On Tue, May 25, 2004 at 03:09:34PM -0700, Jorge Llamb?as wrote:

>

> --- [email protected] wrote:

> > Re: BPFK Section: gadri BTW, xod says that "lo broda poi brode" ==

> > "le broda" (with loss of information, of course).

> >

> > Is that always true, or is it specific to your scheme? Regardless,

> > it seems to be true and I'd appreciate it if you mentioned it

> > somewhere.

>

> {le broda} can refer to anything at all that you have in mind, so I

> don't understand exactly what you mean.

That's OK; it wasn't well thought out.

-Coffee



Posted by Anonymous on Wed 26 of May, 2004 03:00 GMT


> E: {lo nitcu la meripapnz} since Mary Poppins does not exist. {nitcu le

> xamoi archon} which archon also does not exist for all that it is specific

> in my need ("controls the sphere of Mars") Most anything else with

> references to nonexistents will do — unless we have made some unmentioned

> cirumvention.

But that is the same for the proposed {lo}. {mi nitcu lo kriptonite}

has the same kind of problem. That's different than the quantification

issues. When you can meaningfully say {la meripapnz cu prenu}, then you

can also say {mi nitcu la meripapnz}, when not, then not. Same for {le}

and same for {lo}.

> F: Sorry, but Mr. Rabbit (i.e., rabbit b) is not eating grass here and now

> just as he is eating it here and now (as rabbit a).

John is raising his (right) hand here and now, just as he is not

raising his (left) hand here and now.

Rabbit a and rabbit b can't be on the exact same spot at the exact

same time, so Mr Rabbit is eating wherever rabbit a is and he is not

eating wherever rabbit b is.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 04:43 GMT

F: So sumti only apply to existing objects. When the referent does not exist, the bridi containing it is meaningless?? But, of course, we often want sumti in intensional contexts to refer to non-existents. So, even when {la meripapnz prenu} is false or even meaningless, {mi nitcu la meripapnz} may be true (well, at least {mi nitcu tu'a la meripapnz} is and your form is presented as meaning the same as that — or is unexplained). As for quantification, I am not suggesting that {lo broda} means the same as {su'o broda} — indeed I have suggested a range of differences — but only that the inference from {lo} to the fronted {su'o} is valid unless blocked, as it is not in the given cases.

G: Rabbit a is eating here and now, rabbit b is not here now and is not eating. But both rabbit a and rabbit b just ARE Mr. Rabbit, so Mr. Rabbit is both here now and not here now. Note, {lo ractu} refers to Mr. Rabbit, not to some part of Mr. Rabbit; it has the same referent in all occurrences. If you want to change that now, then, of course, you have a metaphysically anomolous reading of the old {lo ractu} — "a part of Mr. Rabbit" rather than "a rabbit" (or "a natural chunk of rabbit goo" or "a manifestation of rabbithood" and so on) but the upshot will be the same in each of these cases, since each occurrence is of a different thing (chunk, part, manifestation, etc.). The talk of Mr.Rabbit then becomes merely useless fluff, since the same work can be done more economically by other means that fit in more naturally with the rest of the language. (The Trobrianders can get away with it since runs throughout their language, doing the work of quantification and abstraction which

appear in Lojban as separatesystems needed elsewhere.)

pc

Jorge Llambías wrote:


> E: {lo nitcu la meripapnz} since Mary Poppins does not exist. {nitcu le

> xamoi archon} which archon also does not exist for all that it is specific

> in my need ("controls the sphere of Mars") Most anything else with

> references to nonexistents will do — unless we have made some unmentioned

> cirumvention.

F: But that is the same for the proposed {lo}. {mi nitcu lo kriptonite}

has the same kind of problem. That's different than the quantification

issues. When you can meaningfully say {la meripapnz cu prenu}, then you

can also say {mi nitcu la meripapnz}, when not, then not. Same for {le}

and same for {lo}.

G:

> F: Sorry, but Mr. Rabbit (i.e., rabbit b) is not eating grass here and now

> just as he is eating it here and now (as rabbit a).

John is raising his (right) hand here and now, just as he is not

raising his (left) hand here and now.

Rabbit a and rabbit b can't be on the exact same spot at the exact

same time, so Mr Rabbit is eating wherever rabbit a is and he is not

eating wherever rabbit b is.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 12:34 GMT


> F: So sumti only apply to existing objects. When the referent does not

> exist, the bridi containing it is meaningless??

Of course not, that's not what I said. I said that lo, le and la

behave in the same way vis-a-vis fiction/non-existence. lo is

not special in this regard.

> But, of course, we often

> want sumti in intensional contexts to refer to non-existents. So, even when

> {la meripapnz prenu} is false or even meaningless, {mi nitcu la meripapnz}

> may be true (well, at least {mi nitcu tu'a la meripapnz} is and your form is

> presented as meaning the same as that — or is unexplained). As for

> quantification, I am not suggesting that {lo broda} means the same as {su'o

> broda} — indeed I have suggested a range of differences — but only that the

> inference from {lo} to the fronted {su'o} is valid unless blocked, as it is

> not in the given cases.

The inference from "I need a box" to "there is some kind of thing

such that I need it" is valid. The inference to "there is some

instance of box such that I need it" is not. It is not in general

valid to infer from the kind to the instances.

> G: Rabbit a is eating here and now, rabbit b is not here now and is not

> eating. But both rabbit a and rabbit b just ARE Mr. Rabbit, so Mr. Rabbit is

> both here now and not here now.

Mr. Rabbit is both here and other-than-here. Kinds can be in more

than one place at the same time. That's not contradictory.

> Note, {lo ractu} refers to Mr. Rabbit, not

> to some part of Mr. Rabbit; it has the same referent in all occurrences.

Right. The same happens with {la djan}. It has the same referent

when I say he is here today and he was not here yesterday. Space

acts for kinds in a similar way to the way time acts for ordinary

individuals.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 14:21 GMT

H: xorxes said <can also say {mi nitcu la meripapnz}, when not, then not. Same for {le}

and same for {lo}.>> It is unclear what this means if not what I tooik it to mean. We do agree, however, that the problems are exactly the same with {la} and {lo} and {le}, which is an improvement on the previous discussion.

I: But the inference from a particular object to the generalization "some object of that sort" holds generally. Thus I take it that you are now saying that {lo broda} refers not to an object but to a kind. I don't think that that position is sustainable without revising the semantics of every word in Lojban — including names and {le} descriptions. And, of course, what I need is not a kiind of thing but a thing of that kind, so the changes merely makes the claim false and leaves us with the problems of saying what we want all over again.

J: Kinds aren't anywhere, manifestations of kinds can each be in only one place at a given time. The two things — kinds and their manifestations — are not to be confused, as they seem to have been here. If predicates are to take kinds as objects then their semantics needs to be revised. But that revision will make them inapplicable to ordinary things, so either we have to double the vocabulary or make all names and {le} etc. sumti about kinds as well. Just learning to use Lojban as written seems a much easier and more natural approach.

K: Well, time does affect individuals differently from space, at least as far as language usually goes — we tend to say that the individual is the same whole over time, but has spatial parts. It is rather hard to build spatial analogs of the time situation for ordinary objects, but temporal analogs for spatial ones are relatively easy: the tomorrow slice of John is here, the today one is not, fits perfectly with John's left hand is raised but his right hand is not. So also, Mr. Rabbit's a manifestation is eating, his b is not. But — unlike the case of John — the references here to Mr.Rabbit play no significant role; the work is all done by the manifestations. It is they that fit in with the rest of the Lojban metaphysics of objects and properties, not of kinds and manifestations (though, of course, we can replicate the results — with a little strain — in that language).

Jorge Llambías wrote:


> F: So sumti only apply to existing objects. When the referent does not

> exist, the bridi containing it is meaningless??

H:Of course not, that's not what I said. I said that lo, le and la

behave in the same way vis-a-vis fiction/non-existence. lo is

not special in this regard.

> But, of course, we often

> want sumti in intensional contexts to refer to non-existents. So, even when

> {la meripapnz prenu} is false or even meaningless, {mi nitcu la meripapnz}

> may be true (well, at least {mi nitcu tu'a la meripapnz} is and your form is

> presented as meaning the same as that — or is unexplained). As for

> quantification, I am not suggesting that {lo broda} means the same as {su'o

> broda} — indeed I have suggested a range of differences — but only that the

> inference from {lo} to the fronted {su'o} is valid unless blocked, as it is

> not in the given cases.

I:The inference from "I need a box" to "there is some kind of thing

such that I need it" is valid. The inference to "there is some

instance of box such that I need it" is not. It is not in general

valid to infer from the kind to the instances.

> G: Rabbit a is eating here and now, rabbit b is not here now and is not

> eating. But both rabbit a and rabbit b just ARE Mr. Rabbit, so Mr. Rabbit is

> both here now and not here now.

J:Mr. Rabbit is both here and other-than-here. Kinds can be in more

than one place at the same time. That's not contradictory.

> Note, {lo ractu} refers to Mr. Rabbit, not

> to some part of Mr. Rabbit; it has the same referent in all occurrences.

K:Right. The same happens with {la djan}. It has the same referent

when I say he is here today and he was not here yesterday. Space

acts for kinds in a similar way to the way time acts for ordinary

individuals.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 14:21 GMT

pc:

> I: But the inference from a particular object to the generalization "some

> object of that sort" holds generally. Thus I take it that you are now saying

> that {lo broda} refers not to an object but to a kind.

{lo broda} does not refer to an instance, that's right. I thought that

was clear from the beginning.

> I don't think that

> that position is sustainable without revising the semantics of every word in

> Lojban — including names and {le} descriptions.

I think it is.

> And, of course, what I need

> is not a kiind of thing but a thing of that kind, so the changes merely makes

> the claim false and leaves us with the problems of saying what we want all

> over again.

When you see John, you are actually seeing a stage of John, but

we don't need to revise {viska} to "a stage of x1 sees a stage

of x2".

> K: Well, time does affect individuals differently from space, at least as far

> as language usually goes — we tend to say that the individual is the same

> whole over time, but has spatial parts.

Similarly Mr Individual is the same whole over instances, just as

John is the same whole over stages.

> It is rather hard to build spatial

> analogs of the time situation for ordinary objects, but temporal analogs for

> spatial ones are relatively easy: the tomorrow slice of John is here, the

> today one is not, fits perfectly with John's left hand is raised but his

> right hand is not. So also, Mr. Rabbit's a manifestation is eating, his b is

> not. But — unlike the case of John — the references here to Mr.Rabbit play

> no significant role; the work is all done by the manifestations.

When you don't care which manifestation is doing the work, all the

reference you need is to Mr.Rabbit.

> It is they

> that fit in with the rest of the Lojban metaphysics of objects and

> properties, not of kinds and manifestations (though, of course, we can

> replicate the results — with a little strain — in that language).

I don't think we need to embed any metaphysics in the language.

The "Mr" talk is just one way of understanding how {lo broda}

behaves logically as a constant term, like {la djan}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 16:01 GMT

L: It is becoming about as clear as anything in all this, but what is not clear is what if anything it does refer to. It is in a place where the norm (I would think) is for reference to individuals, as {le ractu} and {la meripapnz} seem clearly to be, but 9it is not to an individual in any functional sense, apparently, but to an abstraction which then is taken to function as a concrete individual. I think there is some confusion in this notion, so I suppose I do not yet have it right, but nothing said so far has clarified it. Until that is done, I have to say that this section is not yet adequately dealt with, even if the usages are clear (which they are not, for just this reason).

M: I want to see how it will work out: {mi visaka lo ractu} makes {viska} stand for a relation between me and an abstraction, rabbitkind or Mr. Rabbit, while {mi viska le ractu} makes {viska} a relation between me and a concrete object, a selected rabbit. This seems to be two separate meanings of {viska}, marked at best contextually. Lojban literature claims that Lojban predicates are not ambiguous, only vague.

N: A better analogy would surely be that when I see John I see only one side (and maybe not even all of that). But the parts of John — both spatial and temporal — are joiined together in a familiar way, featuring primarily continuity. The parts of Mr. Rabbit lack this feature — among others. Mr.Rabbit appears more like an intermediate abstraction — like a state or a corporation — but lacking the foundation (at least as so far explained) that give these critters legitimacy. I suspect that this can all be corrected and that the notion will have some — maybe even considerable — use. It does not seem to me yet to have anything to do with case like what I need or see or any other fairly normal activity in non-general claims. I am not even clear how it will help in cases like "Cats chase mice" in a way that is clearer than old Lojban devices.

O:But MR.Rabbit is said to be the same over spatially discrete parts, not merely temporal slices and that is markedly different from John: we talk — when it is necessary to avoid confusion or contradiction — about the parts of John, not merely John, but Mr.Rabbit talk is always about Mr. Rabbit simpliciter, not about his parts or manifestations or whatever.

P: When you don't care which manifestation it is, particular quantification — which we already have to have for other reasons — does the job too. Why complicate matters?

Q: But {lo ractu} does not behave like a constant term — or at least you keep refusing to admit ordinary inferences involving constants with respect to it: generalization, negation transparency, apparently subject raising over compounds, and the like do none of them apply to {lo}, but all do to {la}, say.

As I have said, it is probably possible eventually to make a coherent explanation of Mr. Rabbit or whatever, but it seems like a lot of work and it has yet to be demonstrated that the result will solve any real problem — something Lojban does not yet do or do very well. Most of the apparent economies of the notion come about, it seems to me, simply because the notion is still so vague that all manner of very different effects can be attributed to it, even though, were they all actually in it, the result would be incoherent.

Jorge Llambías wrote:

pc:

> I: But the inference from a particular object to the generalization "some

> object of that sort" holds generally. Thus I take it that you are now saying

> that {lo broda} refers not to an object but to a kind.

L:{lo broda} does not refer to an instance, that's right. I thought that

was clear from the beginning.

> I don't think that

> that position is sustainable without revising the semantics of every word in

> Lojban — including names and {le} descriptions.

M:I think it is.

> And, of course, what I need

> is not a kiind of thing but a thing of that kind, so the changes merely makes

> the claim false and leaves us with the problems of saying what we want all

> over again.

N:When you see John, you are actually seeing a stage of John, but

we don't need to revise {viska} to "a stage of x1 sees a stage

of x2".

> K: Well, time does affect individuals differently from space, at least as far

> as language usually goes — we tend to say that the individual is the same

> whole over time, but has spatial parts.

O:Similarly Mr Individual is the same whole over instances, just as

John is the same whole over stages.

> It is rather hard to build spatial

> analogs of the time situation for ordinary objects, but temporal analogs for

> spatial ones are relatively easy: the tomorrow slice of John is here, the

> today one is not, fits perfectly with John's left hand is raised but his

> right hand is not. So also, Mr. Rabbit's a manifestation is eating, his b is

> not. But — unlike the case of John — the references here to Mr.Rabbit play

> no significant role; the work is all done by the manifestations.

P:When you don't care which manifestation is doing the work, all the

reference you need is to Mr.Rabbit.

> It is they

> that fit in with the rest of the Lojban metaphysics of objects and

> properties, not of kinds and manifestations (though, of course, we can

> replicate the results — with a little strain — in that language).

Q:I don't think we need to embed any metaphysics in the language.

The "Mr" talk is just one way of understanding how {lo broda}

behaves logically as a constant term, like {la djan}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 16:01 GMT

An example of an ambiguous sentence with the proposed {lo}

would help clarify matters. From what you say I don't see

how the proposed {lo} is ambiguous.

pc:

> O:But MR.Rabbit is said to be the same over spatially discrete parts, not

> merely temporal slices and that is markedly different from John: we talk --

> when it is necessary to avoid confusion or contradiction — about the parts

> of John, not merely John, but Mr.Rabbit talk is always about Mr. Rabbit

> simpliciter, not about his parts or manifestations or whatever.

Quantification is over the instances. We can talk about them

when we need or want to.

> Q: But {lo ractu} does not behave like a constant term — or at least you

> keep refusing to admit ordinary inferences involving constants with respect

> to it: generalization, negation transparency, apparently subject raising over

> compounds, and the like do none of them apply to {lo}, but all do to {la},

> say.

Generalization to the proper general case does apply to {lo}. It

does not generalize to instances, but then it is not an instance.

Negation transparency does apply to {lo}.

I don't quite understand the third point, but if it applies

to {la} it probably does apply to {lo} as well.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

A brief general discussion to make some context here. One thing cannot both be and not be in the same respect and at the same time. So, when we seem to have a case that contradicts that we have to make a distinction, either in the subject or in the predicate. So, given that John was here yesterday and is not today, we can avoid problems ether by saying "John is here yesterday and is not here today" (taking "John" as a constant directly involved in the situation) or by saying "The yesterday slice of John is here but the today slice is not" (taking John as involved only through his parts, not as a whole. We might of course say "John is such that his yesteday slice is here and his today slice is not" leaving "John" a constant but only his slices doing the work.) The corresponding situations for Mr. Rabbit eating and not are "Mr. Rabbit is a-eating and not b-eating" (or some such thing) and "The a manifestation of Mr. Rabbit is eating but the b-manifestation is not." We can, of

course, generalize on on "John" in any transparent context (and, indeed, in those on "slice of John") but not in opaque contexts, since John might not exist and certainly his yesterday slice might not. Mr Rabbit, on the other hand, as a kind in intension or a property or whatever always exists and so can always be generalized on: to "something" or "some rabbit kind" or.... . But, like slices, manifestations can not be generalized out of opaque contexts. So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

R: See above.

S: But quantification is also over kinds (apparently and I don't mind much) and you seem to want to do both at once — or rather shift back and forth without any indication. In particular, {lo ractu} seems to shift meaning from Mr.Rabbit to a manifestation of Mr Rabbit at your whim. Please give me a rule for figuring out when it is which.

T: What is the proper case: "something" or even "something which is a rabbit kind," I suppose. This clearly goes through. But the interesting case is of a manifestation and there the generalization leaves an reference to manifestations which is not accounted fgor in your examples, so far as I can see. UNless, of course, we are back to the two — undistinguished — meaning of predicates, where the reference to the manifestation is buried in the predicate when it is convenient to do so. The third point is the move from "Mr. Rabbit x's and Mr.Rabbit y's" to Mr. Rabbit x's and y's" Again, this will work (like negation transparency) only if the predicates have been modified to absorb the reference to the different manifestations (and it would be nice to make that absorption explicit at least — aand better, of course, to make it nominal rather than predicative).

pc

Jorge Llambías wrote:

R:An example of an ambiguous sentence with the proposed {lo}

would help clarify matters. From what you say I don't see

how the proposed {lo} is ambiguous.

pc:

> O:But MR.Rabbit is said to be the same over spatially discrete parts, not

> merely temporal slices and that is markedly different from John: we talk --

> when it is necessary to avoid confusion or contradiction — about the parts

> of John, not merely John, but Mr.Rabbit talk is always about Mr. Rabbit

> simpliciter, not about his parts or manifestations or whatever.

S:Quantification is over the instances. We can talk about them

when we need or want to.

> Q: But {lo ractu} does not behave like a constant term — or at least you

> keep refusing to admit ordinary inferences involving constants with respect

> to it: generalization, negation transparency, apparently subject raising over

> compounds, and the like do none of them apply to {lo}, but all do to {la},

> say.

T:Generalization to the proper general case does apply to {lo}. It

does not generalize to instances, but then it is not an instance.

Negation transparency does apply to {lo}.

I don't quite understand the third point, but if it applies

to {la} it probably does apply to {lo} as well.

mu'o mi'e xorxes



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

pc:

> We

> can, of

> course, generalize on on "John" in any transparent context (and, indeed, in

> those on "slice of John") but not in opaque contexts, since John might not

> exist and certainly his yesterday slice might not. Mr Rabbit, on the other

> hand, as a kind in intension or a property or whatever always exists and so

> can always be generalized on: to "something" or "some rabbit kind" or.... .

lo ractu e la djan lenon cu zasti

"Rabbits and John Lennon exist."

lo pavyseljirna e la meripapnz cu na zasti

"Unicorns and Mary Poppins" don't exist.

{lo ractu} works like {la djan lenon}.

{lo pavyseljirna} works like {la meripapnz}.

In this regard there is no lo/la distinction for opaque

or transparent contexts.

> But, like slices, manifestations can not be generalized out of opaque

> contexts. So from "I want Mr. Rabbit" we can presumably generalize to "There

> is something (indeed, a rabbit kind) such that I want it." But "I want Mr.

> Rabbit" does not say the same thing as "I want a rabbit;" that requires "I

> want a manifestation of Mr. Rabbit." And from this we can get "There is

> something (some rabbit kind) such that I want a manifestation of it." My

> problem is to know which of these is what {mi djica lo ractu} means and,

> having said that, how do I say the other.

When we say "I want that rabbit", we don't usually make

the distinction "for that rabbit, I want a (time)slice of it".

Similarly with "I want Mr. Rabbit" we don't have to make

the distinction "for Mr. Rabbit, I want a manifestation

of it."

You can of course still say: {mi djica lo nu mi ponse su'o ractu}

"I want that for some instance of rabbit, I have it", and you can

even shorten it to the somewhat vague {mi djica tu'a su'o ractu}.

You can also say:

lo ractu goi ko'a zo'u mi djica lo nu mi ponse su'o ko'a

Rabbits: I want that there is some instance of them that I have.

But usually you don't need to go that far, just as usually

you don't need to examine what each slice of "that rabbit"

does.

> The situation is easy in old

> Lojban (up to possible

> needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest

> thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica

> da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da

> (poi ka ractu) mi djica tu'a lo ckaji da}.

All of that remains sayable, though I don't really see people

going that route.

> The new {mi djica lo ractu} seems

> to fit somewhere in between — or maybe mix parts of one with parts of the

> other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird

> sense, it seems to me always false since I always have Mr. Rabbit — and all

> other abstractions — as much as I can.)

In what sense can I say to have Mr. Rabbit when I don't have any

of its manifestations?

> S: But quantification is also over kinds (apparently and I don't mind much)

> and you seem to want to do both at once — or rather shift back and forth

> without any indication. In particular, {lo ractu} seems to shift meaning

> from Mr.Rabbit to a manifestation of Mr Rabbit at your whim. Please give me

> a rule for figuring out when it is which.

It is always Mr Rabbit.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

John E Clifford wrote:

>So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

> needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

>

>

There are only 2 actual use cases we must cover.

1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

referring to some real objects; the speaker has revealed all of his

criteria; uses lo)

2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

referring to some real objects, the speaker has not revealed all

criteria but retains some unspoken — he'll let you know if you bring

the wrong rabbit; uses le)

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

xod:

> There are only 2 actual use cases we must cover.

>

> 1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

> referring to some real objects; the speaker has revealed all of his

> criteria; uses lo)

>

> 2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

> referring to some real objects, the speaker has not revealed all

> criteria but retains some unspoken — he'll let you know if you bring

> the wrong rabbit; uses le)

Right.

My only comment is very BTW: I don't like -cfa for this. {cfari}

means "x1 starts to occur", it does not mean "x1 starts at x2"

which seems to be your idea. I suggest {selkaiselfa'a},

"property-oriented" and {kairselfa'a}, "thing-oriented".

Or perhaps {velskiselfa'a}, "description-oriented" and

{selskiselfa'a}, "described-oriented".

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

U: But unlike {la meripapnz na zasti} {lo pavyseljirna na zasti} seems to be false, since Mr. Unicorn exists whether or not any manifestations of him do. Unless, of course, {zasti} has an implicit reference to an instance in it as well or — as now seems likely — all predicates have manifestation references built in. Saying it that way does prejudice the issue a bit, for we could construct a language where the relations with abstractions were primary and either concreta were never mentioned or the predicates with concreta had implicit reduction devices . Depending on the details, these languages are rarely or not at all different from concrete-based languages which raises the question of why bothering to create this elaborate metaphysics . For Lojban, that is; it is perfectly clear Nyaya or Madhyamika might do it, since they need the metaphysics already.

V: Well, in one sense yes we do. That I want when I want a rabbit is an animal with soft fur, long ears, pronounced incisors, etc. That is not Mr. Rabbit, who, as an abstracta, doesn't have ears or incisors or.... Only his manifestations do. All that I want a rabbit doesn't distinguish among is *which* of those manifestations I want.

W: What does the long form say that is different from the short form? It does have the virtue of saying what it means, I think. The fronted case is also not a problem if {lo ractu} refers to the kind or whatever. But then the {su'o lo ractu} does not seem to make much sense, since a kind is not a set or a group of any sort of which we can some members. But in any case, it is an advance since we now actually have to say that it is the instances we want not the kind. And notice, if we do say that, we cannot also say {... ponse lo ractu}, since {ponse} has been identified with a relation having a concreta in its second place (a probably its first as well). {djica} of course has an abstract second place and relates to an (implicit) instance of that abstraction — this time an event one. And we can work these all out eventually.

I guess my ultimate question is "Why bother?" The summary says that this change in {lo} eliminates some problems arising from {lo} = {su'o}, but no examples are given that seem to be problems rather than solecisms. To be sure, {lo} (and probably {su'o} as well) may well have been used in ways that it is not equipped to deal with — kinds and the like, for example. But that means we need more gadri or whatever device we hit upon, not that there is anything wrong with the devices we have for doing their job. Of course, after years of this kind of fiddling, we probably need to be reminded what the jobs of some of these things are and the wiki page would be a good place to start. The only real problems seem to have been dealt with there: the dumb quantifier questions that generate fruitless debate actually extending back further than this one.

X: Well, the long ecxpression has the virtue of being clear and has all the factors explicitly mentioned, including the one hidden in the shorter forms.

Y:The concept exists and I can make the usual uses of it — which is all that having it ever means for concepts.

Z: But you just above said that it was not, I don't have Mr. Rabbit, only his manifestations. Please stick to one side or the other or flag when you are going to shift.

Jorge Llambías wrote:

pc:

> We

> can, of

> course, generalize on on "John" in any transparent context (and, indeed, in

> those on "slice of John") but not in opaque contexts, since John might not

> exist and certainly his yesterday slice might not. Mr Rabbit, on the other

> hand, as a kind in intension or a property or whatever always exists and so

> can always be generalized on: to "something" or "some rabbit kind" or.... .

U:lo ractu e la djan lenon cu zasti

"Rabbits and John Lennon exist."

lo pavyseljirna e la meripapnz cu na zasti

"Unicorns and Mary Poppins" don't exist.

{lo ractu} works like {la djan lenon}.

{lo pavyseljirna} works like {la meripapnz}.

In this regard there is no lo/la distinction for opaque

or transparent contexts.

> But, like slices, manifestations can not be generalized out of opaque

> contexts. So from "I want Mr. Rabbit" we can presumably generalize to "There

> is something (indeed, a rabbit kind) such that I want it." But "I want Mr.

> Rabbit" does not say the same thing as "I want a rabbit;" that requires "I

> want a manifestation of Mr. Rabbit." And from this we can get "There is

> something (some rabbit kind) such that I want a manifestation of it." My

> problem is to know which of these is what {mi djica lo ractu} means and,

> having said that, how do I say the other.

V:When we say "I want that rabbit", we don't usually make

the distinction "for that rabbit, I want a (time)slice of it".

Similarly with "I want Mr. Rabbit" we don't have to make

the distinction "for Mr. Rabbit, I want a manifestation

of it."

W:You can of course still say: {mi djica lo nu mi ponse su'o ractu}

"I want that for some instance of rabbit, I have it", and you can

even shorten it to the somewhat vague {mi djica tu'a su'o ractu}.

You can also say:

lo ractu goi ko'a zo'u mi djica lo nu mi ponse su'o ko'a

Rabbits: I want that there is some instance of them that I have.

But usually you don't need to go that far, just as usually

you don't need to examine what each slice of "that rabbit"

does.

X:> The situation is easy in old

> Lojban (up to possible

> needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest

> thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica

> da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da

> (poi ka ractu) mi djica tu'a lo ckaji da}.

All of that remains sayable, though I don't really see people

going that route.

> The new {mi djica lo ractu} seems

> to fit somewhere in between — or maybe mix parts of one with parts of the

> other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird

> sense, it seems to me always false since I always have Mr. Rabbit — and all

> other abstractions — as much as I can.)

Y:In what sense can I say to have Mr. Rabbit when I don't have any

of its manifestations?

> S: But quantification is also over kinds (apparently and I don't mind much)

> and you seem to want to do both at once — or rather shift back and forth

> without any indication. In particular, {lo ractu} seems to shift meaning

> from Mr.Rabbit to a manifestation of Mr Rabbit at your whim. Please give me

> a rule for figuring out when it is which.

Z:It is always Mr Rabbit.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

Nice — and would that it were so simple. But it fails to cover the case where there are no rabbits or where the one I want doesn't exist, that is the opacity of {djica2}. Other than that, the {lo}-{le} distinction is just fine, as it always has been and without any fluff. In fact, I think that it is perfectly fine even with that problem, provided only that we get over thinking that "I want a rabbit" is {mi djica lo ractu} rather than {mi djica tu'a lo ractu}.

xod wrote: John E Clifford wrote:

>So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

> needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

>

>

There are only 2 actual use cases we must cover.

1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

referring to some real objects; the speaker has revealed all of his

criteria; uses lo)

2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

referring to some real objects, the speaker has not revealed all

criteria but retains some unspoken — he'll let you know if you bring

the wrong rabbit; uses le)

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Wed 26 of May, 2004 22:27 GMT

Nice — and would that it were so simple. But it fails to cover the case where there are no rabbits or where the one I want doesn't exist, that is the opacity of {djica2}. Other than that, the {lo}-{le} distinction is just fine, as it always has been and without any fluff. In fact, I think that it is perfectly fine even with that problem, provided only that we get over thinking that "I want a rabbit" is {mi djica lo ractu} rather than {mi djica tu'a lo ractu}.

xod wrote:John E Clifford wrote:

>So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

> needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

>

>

There are only 2 actual use cases we must cover.

1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

referring to some real objects; the speaker has revealed all of his

criteria; uses lo)

2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

referring to some real objects, the speaker has not revealed all

criteria but retains some unspoken — he'll let you know if you bring

the wrong rabbit; uses le)

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Wed 26 of May, 2004 22:28 GMT


> U: But unlike {la meripapnz na zasti} {lo pavyseljirna na zasti} seems to be

> false, since Mr. Unicorn exists whether or not any manifestations of him do.

Mr. Unicorn is as inexistent as Mary Poppins.

{lo pavyseljirna na zasti} simply means "Unicorns don't exist"

in the same way that {la meripapnz na zasti} means "Mary Poppins

does not exist".

To understand the meaning of {lo pavyseljirna na zasti} you need

to know what {lo pavyseljirna} means in the same way that you need

to know what {la meripapnz} means to understand {la meripapnz na zasti}.

That the meanings of both those expressions "exist" doesn't imply that

there are real objects in the world that respond to them.

There is no distinction in principle between Mary Poppins and

Mr Unicorn.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 26 of May, 2004 23:36 GMT

John E Clifford wrote:

>Nice — and would that it were so simple. But it fails to cover the case where there are no rabbits or where the one I want doesn't exist, that is the opacity of {djica2}. Other than that, the {lo}-{le} distinction is just fine, as it always has been and without any fluff.

>

By "lo-le" distinction I hope you mean the one I've described below.

It does in fact cover the case of no rabbits. Questions of existence are

a distraction. There is no existence claim in lo here. And even if all

doctors are slaughtered, it will still be possible to ask for one and be

understood, even if the listeners are powerless to help. Therefore such

sentences do express something which is actually invariant with respect

to the existence of doctors.

> In fact, I think that it is perfectly fine even with that problem, provided only that we get over thinking that "I want a rabbit" is {mi djica lo ractu} rather than {mi djica tu'a lo ractu}.

>

>

We do intend {mi djica lo ractu} to in fact gloss as {I want a rabbit}.

>xod wrote:John E Clifford wrote:

>

>

>

>>So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

>>needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

>>

>>

>>

>>

>

>There are only 2 actual use cases we must cover.

>

>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

>referring to some real objects; the speaker has revealed all of his

>criteria; uses lo)

>

>2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

>referring to some real objects, the speaker has not revealed all

>criteria but retains some unspoken — he'll let you know if you bring

>the wrong rabbit; uses le)

>

>

>

>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 27 of May, 2004 15:41 GMT

Well, there are no manifestations of Mr. Unicorn, but Mr. Unicorn exists like all kinds. You really have to settle down on what this locution means — or, I suspect, get better about saying what that is. If Mr. Unicorn does not exist, then he is only a disguised way of talking about his instances and so in practice not different from "some unicorn." The juggling to get something externally generalizable out of such expressions is also then just a trick that needlessly — and misleadingly — hides what is going on: the generalization on, e.g., {le ka pavyseljirna} (I do agree that the gadri there is otiose, even odious).

pc

Jorge Llambías wrote:


> U: But unlike {la meripapnz na zasti} {lo pavyseljirna na zasti} seems to be

> false, since Mr. Unicorn exists whether or not any manifestations of him do.

Mr. Unicorn is as inexistent as Mary Poppins.

{lo pavyseljirna na zasti} simply means "Unicorns don't exist"

in the same way that {la meripapnz na zasti} means "Mary Poppins

does not exist".

To understand the meaning of {lo pavyseljirna na zasti} you need

to know what {lo pavyseljirna} means in the same way that you need

to know what {la meripapnz} means to understand {la meripapnz na zasti}.

That the meanings of both those expressions "exist" doesn't imply that

there are real objects in the world that respond to them.

There is no distinction in principle between Mary Poppins and

Mr Unicorn.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 15:41 GMT

But asking for a doctor is just the sort of opaque reference where — as you say — the real existence is irrelevant. Reporting what a doctor did when there are no doctors is another matter and one where existence is relevant.

As for {mi djica lo ractu}, I can think of several ways whereby this would be a meaningful way to say I want a rabbit. Unfortunately, none of them seem to be the way that {lo} is presently (incoherently) explained. And, of course, almost all of them leave problems problems in other places (contrasts with {le} in some contexts, permissible external references, and the like) which are avoided by the old Lojban locution. Can you — xorxes having so far not — provide a clear case of properly used old {lo} which causes a problem which new {lo} solves?

xod wrote:

John E Clifford wrote:

>Nice — and would that it were so simple. But it fails to cover the case where there are no rabbits or where the one I want doesn't exist, that is the opacity of {djica2}. Other than that, the {lo}-{le} distinction is just fine, as it always has been and without any fluff.

>

By "lo-le" distinction I hope you mean the one I've described below.

It does in fact cover the case of no rabbits. Questions of existence are

a distraction. There is no existence claim in lo here. And even if all

doctors are slaughtered, it will still be possible to ask for one and be

understood, even if the listeners are powerless to help. Therefore such

sentences do express something which is actually invariant with respect

to the existence of doctors.

> In fact, I think that it is perfectly fine even with that problem, provided only that we get over thinking that "I want a rabbit" is {mi djica lo ractu} rather than {mi djica tu'a lo ractu}.

>

>

We do intend {mi djica lo ractu} to in fact gloss as {I want a rabbit}.

>xod wrote:John E Clifford wrote:

>

>

>

>>So from "I want Mr. Rabbit" we can presumably generalize to "There is something (indeed, a rabbit kind) such that I want it." But "I want Mr. Rabbit" does not say the same thing as "I want a rabbit;" that requires "I want a manifestation of Mr. Rabbit." And from this we can get "There is something (some rabbit kind) such that I want a manifestation of it." My problem is to know which of these is what {mi djica lo ractu} means and, having said that, how do I say the other. The situation is easy in old Lojban (up to possible

>>needed predicates: {mi djica tu'a le ka ractu} (taking {ka}as the nearest thing to -kind, for the moment) whence {su'o da (poi ka ractu) zo'u mi djica da} and {mi djica tu'a lo ractu}, whence, by a longer route perhaps, {su'o da (poi ka ractu) mi djica tu'a lo ckaji da}. The new {mi djica lo ractu} seems to fit somewhere in between — or maybe mix parts of one with parts of the other. (Incidentally, if "I want Mr. Rabbit" is not meant in some weird sense, it seems to me always false since I always have Mr. Rabbit — and all other abstractions — as much as I can.)

>>

>>

>>

>>

>

>There are only 2 actual use cases we must cover.

>

>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaicfa;

>referring to some real objects; the speaker has revealed all of his

>criteria; uses lo)

>

>2. I want the rabbit, a certain rabbit, not any other. (kaicfa;

>referring to some real objects, the speaker has not revealed all

>criteria but retains some unspoken — he'll let you know if you bring

>the wrong rabbit; uses le)

>

>

>

>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 27 of May, 2004 15:41 GMT

pc:

> You really have to settle down on what this locution means — or,

> I suspect, get better about saying what that is.

I'll try. {lo broda} is, first and foremost, a constant term,

just like {la brod.} is a constant term.

lo ractu can be seen and touched. That means "rabbits can be seen

and touched".

la ract can be seen and touched. That means "Rasht can be seen and

touched".

When you see or touch an instance of lo ractu, you are, in the

very same act, seeing or touching lo ractu.

When you see or touch a stage of la ract, you are, in the very

same act, seeing or touching la ract.

(A "stage" differs from a "slice" in that a stage has some time depth,

so a stage is a series of contiguous slices. The analogy of instances

with stages seems better than the analogy with slices, which, having

zero time depth, are somewhat unreal. The best analogy is perhaps that

of stages of individuals with instances of substances, because two

time-contiguous stages give you one longer stage, two space-contiguous

instances give you one bigger instance of the substance.)

lo ractu cu zasti. Rabbits are real/actual.

la ract cu zasti. Rasht is real/actual.

For the above to be true, it is necessary that some instance

of lo ractu and some stage of la ract are real/actual, too. That's

just how {zasti} works. You can't zasti if you don't have some

instance/stage that zasti.

lo pavyseljirna cu xanri. Unicorns are imaginary.

la pavyseljirn cu xanri. Pavyseljirn is imaginary.

Imaginary things don't have real instances/stages. They may

or may not have imaginary instances/stages, depending on how

elaborately they are imagined.

> The juggling to get something

> externally generalizable out of such expressions is also then just a trick

> that needlessly — and misleadingly — hides what is going on: the

> generalization on, e.g., {le ka pavyseljirna} (I do agree that the gadri

> there is otiose, even odious).

The "Mr" talk is one way of conceptualizing things. As long as

{lo broda} behaves like a constant term (i.e. it is transparent

to negation, it can be repeated with anaphora, etc) then it

doesn't matter much how one conceptualizes it. That such a thing

is needed is evident to anyone who has used the language to some

extent. It is extremely cumbersome to reduce all your claims to

instances when in very many cases the distribution of instances

is obvious or irrelevant. When we do want to go into instances,

all the usual machinery remains available: we simply quantify

over all of them with {PA lo broda}, or refer to a specific

instance or group of instances directly with le/la.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 15:41 GMT

pc (to xod):

> Can you — xorxes having so far not

> — provide a clear case of properly used old {lo} which causes a problem

> which new {lo} solves?

But properly used old {lo} never causes a problem! It corresponds

exactly to both old and proposed {su'o lo}, which is unproblematic.

What the new {lo} does is allow things that propoerly used old {lo}

cannot handle easily or at all.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:27 GMT

A': I get that it is a constant term, i.e., refers to the same thing in all contexts. The issue is what that thing is. Apparently it is a concretum and so the talk about it as an abstractum is wrong, in spite of its making sense of the constancy, which seems to be lacking in the concrete sense.

B: So lo ractu is just the sum (collective? not set surely) of all its "instances", that is of all rabbits. It has, then, (and this may distinguish it from collectives, which might only have the cooperative properties of more than one thing) all the properties of all the rabbits, in some way that does not create contradictions.

I think my basic problem here is that you are trying to explain a synthetic category (like a collective or so) by analogy with an analytic category like a slicew or stage. with analytic categories you can fall back on the original object being analysed to guarantee and at least partially explain what holds the pieces together, with a synthetic category we have to provide that glue explicitly and this has not been done. Why should I (starting from where I am) see all these individuals (rabbits in this case) as being instances of something else concrete. This is not even like seeing all nature as One or even treating an ecosystem as an entity, for what we are dealing with has no "natural" cohesion. What is needed then is a convincing artificial one and that isn't here yet. (This is not to say that, starting from a different perceptual framework — a Madhyamika Buddhis, say — I wouldn't find this natural as well, though I am not sure that they actually would).

C: Assuming that {la ract} is meant to be a name for an ordinary thing — a guy called Bunny, for example — then more than a stage has to exist for it to exist. The stage has to fit into a continuous series of contiguous stages satisfying an array of further condition. Otherwise, Bunny falls into some category like delusions or illusions or ...

Mr. Rabbit on the other hand seems to be exactly nothing other than a bunch of "instances" with no yet explained further conditions. And as such it seems pointless, given that we have the instances. So, in addition to the glue, we need a raison d'etre for this notion.

D: Thse, of course, raise the old paradox of how to say of something that does not exist that it does not exist. I don't see that Mr. Unicorn helps here at all (though short-scope {lo} would if we take "imaginary" to be world-creating — which we should, for a variety of other reasons as well).

E: Frinstance? Lojban is almost always going to seem cumbersome to speakers of natural language because it has a built in precising mechanism and has not yet developed good conventions for work-arounds. I would assume that people who use tha language a lot (a class which is almost coextensive with you) have begun to develop those things. But that is very different from changing the basics of the language, which is what new {lo} seems to do.

What exactly is the advantage of making {lo ractu} a constant, when the phenomena being described involve variable references? As for the obvious/irrelevant distribution, that is exactly what particular quantifiers do.

Jorge Llambías wrote:

pc:

> You really have to settle down on what this locution means — or,

> I suspect, get better about saying what that is.

A':I'll try. {lo broda} is, first and foremost, a constant term,

just like {la brod.} is a constant term.

lo ractu can be seen and touched. That means "rabbits can be seen

and touched".

la ract can be seen and touched. That means "Rasht can be seen and

touched".

B:When you see or touch an instance of lo ractu, you are, in the

very same act, seeing or touching lo ractu.

When you see or touch a stage of la ract, you are, in the very

same act, seeing or touching la ract.

(A "stage" differs from a "slice" in that a stage has some time depth,

so a stage is a series of contiguous slices. The analogy of instances

with stages seems better than the analogy with slices, which, having

zero time depth, are somewhat unreal. The best analogy is perhaps that

of stages of individuals with instances of substances, because two

time-contiguous stages give you one longer stage, two space-contiguous

instances give you one bigger instance of the substance.)

C:lo ractu cu zasti. Rabbits are real/actual.

la ract cu zasti. Rasht is real/actual.

For the above to be true, it is necessary that some instance

of lo ractu and some stage of la ract are real/actual, too. That's

just how {zasti} works. You can't zasti if you don't have some

instance/stage that zasti.

D:lo pavyseljirna cu xanri. Unicorns are imaginary.

la pavyseljirn cu xanri. Pavyseljirn is imaginary.

Imaginary things don't have real instances/stages. They may

or may not have imaginary instances/stages, depending on how

elaborately they are imagined.

> The juggling to get something

> externally generalizable out of such expressions is also then just a trick

> that needlessly — and misleadingly — hides what is going on: the

> generalization on, e.g., {le ka pavyseljirna} (I do agree that the gadri

> there is otiose, even odious).

E:The "Mr" talk is one way of conceptualizing things. As long as

{lo broda} behaves like a constant term (i.e. it is transparent

to negation, it can be repeated with anaphora, etc) then it

doesn't matter much how one conceptualizes it. That such a thing

is needed is evident to anyone who has used the language to some

extent. It is extremely cumbersome to reduce all your claims to

instances when in very many cases the distribution of instances

is obvious or irrelevant. When we do want to go into instances,

all the usual machinery remains available: we simply quantify

over all of them with {PA lo broda}, or refer to a specific

instance or group of instances directly with le/la.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:27 GMT

F: Frinstance again? Are these things that old {lo} should do, rather than some other (perhaps new) gadri or other device. Were you supplementing {lo} rather than replacing it, there would probably be no problems here, but replacing it wihtout ground or reason seem hubristic as well as unnecessary.

Jorge Llambías wrote:

pc (to xod):

> Can you — xorxes having so far not

> — provide a clear case of properly used old {lo} which causes a problem

> which new {lo} solves?

But properly used old {lo} never causes a problem! It corresponds

exactly to both old and proposed {su'o lo}, which is unproblematic.

F:What the new {lo} does is allow things that propoerly used old {lo}

cannot handle easily or at all.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:28 GMT

pc:

> I think my basic problem here is that you are trying to explain a synthetic

> category (like a collective or so) by analogy with an analytic category like

> a slicew or stage.

The analogy is kind:instance::individual:stage

so the kind is analogous to the individual and the

instance to the stage.

> with analytic categories you can fall back on the

> original object being analysed to guarantee and at least partially explain

> what holds the pieces together, with a synthetic category we have to provide

> that glue explicitly and this has not been done.

The glue is the property in question. lo ka ractu is the

property that all the instances of lo ractu have in common.

> Why should I (starting from

> where I am) see all these individuals (rabbits in this case) as being

> instances of something else concrete.

Because they share a common name/description?

> This is not even like seeing all

> nature as One or even treating an ecosystem as an entity, for what we are

> dealing with has no "natural" cohesion. What is needed then is a convincing

> artificial one and that isn't here yet.

I offer their description, "ractu", or the property they all share,

{lo ka ractu}.

> (This is not to say that, starting

> from a different perceptual framework — a Madhyamika Buddhis, say — I

> wouldn't find this natural as well, though I am not sure that they actually

> would).

>

> C: Assuming that {la ract} is meant to be a name for an ordinary thing — a

> guy called Bunny, for example — then more than a stage has to exist for it

> to exist. The stage has to fit into a continuous series of contiguous stages

> satisfying an array of further condition.

Intermittent existence is ruled out even in principle, by definition?

{da ru'inai zasti} is false by definition?

> Otherwise, Bunny falls into some

> category like delusions or illusions or ...

If it has at least one stage that exists, I bet it has to exist.

> Mr. Rabbit on the other hand seems to be exactly nothing other than a bunch

> of "instances" with no yet explained further conditions. And as such it

> seems pointless, given that we have the instances. So, in addition to the

> glue, we need a raison d'etre for this notion.

The glue is that the instances share a description, the raison

d'etre is simplicity in use (constants are much easier to handle than

quantified terms).

> D: Thse, of course, raise the old paradox of how to say of something that

> does not exist that it does not exist. I don't see that Mr. Unicorn helps

> here at all (though short-scope {lo} would if we take "imaginary" to be

> world-creating — which we should, for a variety of other reasons as well).

In any case, the paradox is neither more nor less paradoxical for

Mr. Unicorn than for Ms. Poppins.

> E: Frinstance? Lojban is almost always going to seem cumbersome to speakers

> of natural language because it has a built in precising mechanism and has not

> yet developed good conventions for work-arounds. I would assume that people

> who use tha language a lot (a class which is almost coextensive with you)

> have begun to develop those things. But that is very different from changing

> the basics of the language, which is what new {lo} seems to do.

> What exactly is the advantage of making {lo ractu} a constant, when the

> phenomena being described involve variable references? As for the

> obvious/irrelevant distribution, that is exactly what particular quantifiers

> do.

But not always the references are variable. When I talk of rabbits

in general, I am talking of one thing: rabbits, not about some

rabbit or each rabbit. Examples of things that the old {lo} is not

well equipped to handle are most of the lo examples under the

proposed definition.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:28 GMT

F:But the disanalogy is stronger-- one is the product of analysis, the other of synthesis, which require very different sorts of things. We know that the totality exists and have doubts about the stages in the one case; we jknow that the instances exist but have doubts about the totality in the other.

G: Te property of being a rabbit holds together the stages of individual rabbits, but it does not (or at least has not been shown to) hld together various rabbits into a new entity.

H: But I have no reason (that I know of or have been informed of) to think that all the men named John form a concrete entity, nor for that matter any other things which share a name or even a description. that they form a variety of *abstract* entities (in some sense of "form") I have no doubt (which is probably why I — trying my damnedest to be fair and make sense out of what you say — keep coming back to Mr. Rabbit as an abstract).

I:There are things (I'm not sure how concrete they are) that have intermittent existences — Poland springs to mind, but they are held together by quite a bit more than the name (history, culture, language, emotion, even a constitution in some cases — though not Poland's). I should not think that Mr. Rabbit — or Mr.anything else concrete would be such an intermediate case. (The official Christian view of the afterlife, bodily resurrection, is an interesting case, too.)

J: Maybe, but as noted, not as the sort of thing it seems to be.

K: Hey, a reasonably clear statement of why do this: it is simpler. Doing a way with quantifiers altogether and using only constants (it can be done) would be simpler still, it would seem. But there remains the problem of 1) explaining what the heck these constants mean and 2)working out the screamingly difficult connection between what are linguistically constants and what are logically quantified variables. Part of the point of Lojban, recall is that the logical operations are to be near the surface, not buried — at least in langue, even if not always in parole. Since we had a smoothly functioning device for doing this, why replace it with one that creates a mass of headaches by burying the logic away in references to mythic objects (which, it turns out, we cannot in fact refer to since every use of the supposed referring expression turns out to refer in fact to an instance for the practical semantic moves, e.g., finding out whether a sentence using the expression is true).

L: But the paradox disappears completely with quantification (one of the reasons for its development, in fact). {su'o pavyseljirna na zasti}

M: But why do you think that {lo ractu} old style would be used to say something about the class of rabbits or the collective of rabbits or whatever it is that you seem to think "rabbits" refers to in English/ {lo} makes a lousy tense marker, too, but that is not a reason to replace it; it is a reason to get a tense marker that does what is wanted. (Of coure, I tend to think that most of the kinds of thins you are talking about are malglicoisms in spades, but I am willing to think there may be a residue that need work: more gadri {le'e} and {lo'e}, which refer not to individuals but create a summary of claims about the critters of the appropriate sort: "rabbits," if it is not just {lo'e ractu}, is clearly in the same area. Note that it does not refer to some mythic individual — or to any describable combination of reals ones and so is not quantifiable, either in itself nor as a base for generalization. I forget what the gadri page says about this.)

If what you say about the {lo} examples — I have to admit I stopped at {... nitcu lo ...}, then the gadri page is in worse shape than even I thought, since, by your descripotion, it gives as exemplary cases of {lo} things that are not (clearly, generally agreed) cases of {lo} at all.

Byt the way, somehwere earlier you called {tu'a lo ractu} vague. It is, of course, since it omits information about what I want/need/etc. for or what I dreamed about it and so on. But this is marked less vague than {lo ractu} in your usage, which also omits all that information (and does not even indicate that it may be significant) but plunks us down with something that is not even a rabbit (or is only in an indirect sort of way).

Jorge Llambías wrote:

pc:

> I think my basic problem here is that you are trying to explain a synthetic

> category (like a collective or so) by analogy with an analytic category like

> a slicew or stage.

F:The analogy is kind:instance::individual:stage

so the kind is analogous to the individual and the

instance to the stage.

> with analytic categories you can fall back on the

> original object being analysed to guarantee and at least partially explain

> what holds the pieces together, with a synthetic category we have to provide

> that glue explicitly and this has not been done.

G:The glue is the property in question. lo ka ractu is the

property that all the instances of lo ractu have in common.

> Why should I (starting from

> where I am) see all these individuals (rabbits in this case) as being

> instances of something else concrete.

H:Because they share a common name/description?

> This is not even like seeing all

> nature as One or even treating an ecosystem as an entity, for what we are

> dealing with has no "natural" cohesion. What is needed then is a convincing

> artificial one and that isn't here yet.

I offer their description, "ractu", or the property they all share,

{lo ka ractu}.

> (This is not to say that, starting

> from a different perceptual framework — a Madhyamika Buddhis, say — I

> wouldn't find this natural as well, though I am not sure that they actually

> would).

>

> C: Assuming that {la ract} is meant to be a name for an ordinary thing — a

> guy called Bunny, for example — then more than a stage has to exist for it

> to exist. The stage has to fit into a continuous series of contiguous stages

> satisfying an array of further condition.

I:Intermittent existence is ruled out even in principle, by definition?

{da ru'inai zasti} is false by definition?

> Otherwise, Bunny falls into some

> category like delusions or illusions or ...

J:If it has at least one stage that exists, I bet it has to exist.

> Mr. Rabbit on the other hand seems to be exactly nothing other than a bunch

> of "instances" with no yet explained further conditions. And as such it

> seems pointless, given that we have the instances. So, in addition to the

> glue, we need a raison d'etre for this notion.

K:The glue is that the instances share a description, the raison

d'etre is simplicity in use (constants are much easier to handle than

quantified terms).

> D: Thse, of course, raise the old paradox of how to say of something that

> does not exist that it does not exist. I don't see that Mr. Unicorn helps

> here at all (though short-scope {lo} would if we take "imaginary" to be

> world-creating — which we should, for a variety of other reasons as well).

L:In any case, the paradox is neither more nor less paradoxical for

Mr. Unicorn than for Ms. Poppins.

> E: Frinstance? Lojban is almost always going to seem cumbersome to speakers

> of natural language because it has a built in precising mechanism and has not

> yet developed good conventions for work-arounds. I would assume that people

> who use tha language a lot (a class which is almost coextensive with you)

> have begun to develop those things. But that is very different from changing

> the basics of the language, which is what new {lo} seems to do.

> What exactly is the advantage of making {lo ractu} a constant, when the

> phenomena being described involve variable references? As for the

> obvious/irrelevant distribution, that is exactly what particular quantifiers

> do.

M:But not always the references are variable. When I talk of rabbits

in general, I am talking of one thing: rabbits, not about some

rabbit or each rabbit. Examples of things that the old {lo} is not

well equipped to handle are most of the lo examples under the

proposed definition.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:28 GMT

John E Clifford wrote:

>But asking for a doctor is just the sort of opaque reference where — as you say — the real existence is irrelevant. Reporting what a doctor did when there are no doctors is another matter and one where existence is relevant.

>

>

It is interesting that existence has historically plagued lo, when we

see it is really more of a factor for le.

>

>As for {mi djica lo ractu}, I can think of several ways whereby this would be a meaningful way to say I want a rabbit. Unfortunately, none of them seem to be the way that {lo} is presently (incoherently) explained.

>

Perhaps then you should try working backwards from the premise: the

proposed lo is intended to work in that sentence as "a rabbit".

> And, of course, almost all of them leave problems problems in other places (contrasts with {le} in some contexts, permissible external references, and the like) which are avoided by the old Lojban locution.

>

Can you offer concrete examples of these problems?

> Can you — xorxes having so far not — provide a clear case of properly used old {lo} which causes a problem which new {lo} solves?

>

>

An end to erroneous idea that lo provides an existence claim and a

secret "there is a" phrase, and the ability to use lo mikce for "any

doctor" without cmavo torture and a fortnight of heated discussion each

time. Paraphrasing Robin: any language in which you cannot ask for a

doctor is a toy!

Once again, and if you have concrete counterexample use cases, I'd be

interested to see them:

>>There are only 2 actual use cases we must cover.

>>

>>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaisanji;

>>referring to some real objects; the speaker has revealed all of his

>>criteria; uses lo)

>>

>>2. I want the rabbit, a certain rabbit, not any other. (kairsanji;

>>referring to some real objects, the speaker has not revealed all

>>criteria but retains some unspoken — he'll let you know if you bring

>>the wrong rabbit; uses le)

>>

>>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 27 of May, 2004 21:49 GMT

pc:

> F:But the disanalogy is stronger-- one is the product of analysis, the other

> of synthesis, which require very different sorts of things. We know that the

> totality exists and have doubts about the stages in the one case; we jknow

> that the instances exist but have doubts about the totality in the other.

I don't have more doubts about the stages than the individual, or

about the kind than the instances. They are all artifacts of language

to me. Maybe that's why we can't seem to understand each other.

> L: But the paradox disappears completely with quantification (one of the

> reasons for its development, in fact). {su'o pavyseljirna na zasti}

What about {su'o pavyseljirna cu xanri}?

> M: But why do you think that {lo ractu} old style would be used to say

> something about the class of rabbits or the collective of rabbits or whatever

> it is that you seem to think "rabbits" refers to in English/ {lo} makes a

> lousy tense marker, too, but that is not a reason to replace it; it is a

> reason to get a tense marker that does what is wanted.

{lo} was redundant in its function, being equivalent to {su'o},

so it was the logical choice. Also, there's the historical conexion

to Loglan {lo}. Also, the simplest gadri should be the most general

one. {lo} is to gadri as {cu} is to tenses.

> If what you say about the {lo} examples — I have to admit I stopped at {...

> nitcu lo ...}, then the gadri page is in worse shape than even I thought,

> since, by your descripotion, it gives as exemplary cases of {lo} things that

> are not (clearly, generally agreed) cases of {lo} at all.

They are examples of the proposed lo, of course. This is the adrees

of the page, in case you want to discuss actual examples:

http://www.lojban.org/tiki/BPFK+Section%3A+gadri

> Byt the way, somehwere earlier you called {tu'a lo ractu} vague. It is, of

> course, since it omits information about what I want/need/etc. for

Both {djica} and {nitcu} have an x3 for "for", so the vagueness

of tu'a is of a different sort. For example:

mi nitcu tu'a lo tanxe lo nu mi punji lo cukta ty(?)

I need (something about) a box so that I put some books in it(?)

The "something" is probably "having it". I'm not sure if the {ty}

is correct here. Can you refer to a quantified something that's

inside a different abstraction?

> dreamed about it and so on. But this is marked less vague than {lo ractu} in

> your usage, which also omits all that information (and does not even indicate

> that it may be significant) but plunks us down with something that is not

> even a rabbit (or is only in an indirect sort of way).

lo ractu does, of course, ractu. {mi nitcu lo tanxe lo nu mi punji

lo cukta ty} is simply "I need a box to put books in it", no more vague

than that.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 27 of May, 2004 21:49 GMT

  • Why more {le} than {lo}? Both claim existence, though in different way. But then I don't see it as a problem for {lo} even.
  • that is my premise (or rather the premise of earlier Lojban): {lo} is equivalent to a particular quantifier (except perhaps distributionally). And it is that in all contexts. In event clauses (at least) its range is limited to the world of the mentioned event, which may be different from the existents outside in this world. This fact keeps us from carrying that quantifier outward ({tu'a} is merely a shorthand for an abstract description and so has the same effect). The problem with {mi djica lo ractu} is that it looks like that reference can be fronted as though it certainly referred to something in this world, and aparently it can be with new {lo}, with all manner of strange consequences (changing true to false as far as I can see).
  • Not probably without developing the various ways of reading things. But briefly, (about {le}) if we make {lo ractu} really about a single thing, then all predicates to which it can be attached have to contain a reference to a manifestation of that thing, which then will conmflict with a {le} sumti, which presumably does not refer to something that has manifestations. So, every predicate becomes ambiguous. And of course (about distribution), even if {mi djica lo pavyseljirna} is true — as it well may be (despite xorxes saying thing which seem to require it to be false). {lo pavyseljirna goi ko'a z'u mi djica ko'a} is clearly false, since there are no unicorns.
  • How are these problems? The first is just part of what "a" basically means in English and has carried over virtually unchanged to Lojban. The second is (if I understand the case) nothing to do with {lo} at all and so can't be counted against it. Further, there is no evidence that the new {lo} as explained does anything to help this situation. Depending on the context, "any doctor" would seem to be either {lo mikce} (old sense) or {ro mikce}.

And, of course there is not problem in asking for a doctor with the old sense of of {lo} (at worst you have to remember that imperatives are intensional contexts, but that is obvious).. The one problem that I have ever seen that might occasionally cause trouble for us learners is theat of negation-transparency and solutions to that can be made up within old {lo} at minimal cost. And even without a change, the solution is just to get used to thinking about what negations do and speaking accordingly.

  • Several mentioned in passing here and others in the xorxes notes. These may not be real counterexamples to new {lo} or they may be proofs that new {lo} is incoherent. At the moment I am torn, but come down of the middle position that, even if new {lo} is coherent, it is an unnecesssary change, a misguided (probably malglico) attempt to solve some problems basically unrelated to {lo}.

xod wrote:

John E Clifford wrote:

>But asking for a doctor is just the sort of opaque reference where — as you say — the real existence is irrelevant. Reporting what a doctor did when there are no doctors is another matter and one where existence is relevant.

>

>

  • It is interesting that existence has historically plagued lo, when we

see it is really more of a factor for le.

>

>As for {mi djica lo ractu}, I can think of several ways whereby this would be a meaningful way to say I want a rabbit. Unfortunately, none of them seem to be the way that {lo} is presently (incoherently) explained.

>

  • Perhaps then you should try working backwards from the premise: the

proposed lo is intended to work in that sentence as "a rabbit".

> And, of course, almost all of them leave problems problems in other places (contrasts with {le} in some contexts, permissible external references, and the like) which are avoided by the old Lojban locution.

>

  • Can you offer concrete examples of these problems?

> Can you — xorxes having so far not — provide a clear case of properly used old {lo} which causes a problem which new {lo} solves?

>

>

  • An end to erroneous idea that lo provides an existence claim and a

secret "there is a" phrase, and the ability to use lo mikce for "any

doctor" without cmavo torture and a fortnight of heated discussion each

time. Paraphrasing Robin: any language in which you cannot ask for a

doctor is a toy!

  • Once again, and if you have concrete counterexample use cases, I'd be

interested to see them:

>>There are only 2 actual use cases we must cover.

>>

>>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaisanji;

>>referring to some real objects; the speaker has revealed all of his

>>criteria; uses lo)

>>

>>2. I want the rabbit, a certain rabbit, not any other. (kairsanji;

>>referring to some real objects, the speaker has not revealed all

>>criteria but retains some unspoken — he'll let you know if you bring

>>the wrong rabbit; uses le)

>>

>>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 27 of May, 2004 21:49 GMT

John E Clifford wrote:

> * Why more {le} than {lo}? Both claim existence, though in different

> way. But then I don't see it as a problem for {lo} even.

In kairsanji, existence is implied because the thing existed at least

when the speaker made its acquaintance. In selkaisanji, existence is not

implied at all; the speaker is only discussing characteristics.

> ** that is my premise (or rather the premise of earlier Lojban): {lo}

> is equivalent to a particular quantifier (except perhaps

> distributionally). And it is that in all contexts. In event clauses

> (at least) its range is limited to the world of the mentioned event,

> which may be different from the existents outside in this world. This

> fact keeps us from carrying that quantifier outward ({tu'a} is merely

> a shorthand for an abstract description and so has the same effect).

> The problem with {mi djica lo ractu} is that it looks like that

> reference can be fronted as though it certainly referred to something

> in this world,

What does "fronted" mean? Please give an example of the problem it poses.

> and aparently it can be with new {lo}, with all manner of strange

> consequences (changing true to false as far as I can see). ***Not

> probably without developing the various ways of reading things. But

> briefly, (about {le}) if we make {lo ractu} really about a single

> thing, then all predicates to which it can be attached have to contain

> a reference to a manifestation of that thing, which then will

> conmflict with a {le} sumti, which presumably does not refer to

> something that has manifestations. So, every predicate becomes ambiguous.

I don't follow this at all.

> And of course (about distribution), even if {mi djica lo pavyseljirna}

> is true — as it well may be (despite xorxes saying thing which seem

> to require it to be false). {lo pavyseljirna goi ko'a z'u mi djica

> ko'a} is clearly false, since there are no unicorns.

Why would this be more false than with the substitution of ko'a with its

value?

> ****How are these problems? The first is just part of what "a"

> basically means in English and has carried over virtually unchanged to

> Lojban.

I was talking about the existence claim. Does English "a" have one?

> The second is (if I understand the case) nothing to do with {lo} at

> all and so can't be counted against it.

If you read the description in the Book, you'll find it comes very close

to what Jorge is proposing, and that at one time, the idea of using lo

for "a doctor" would not have been controversial. I don't know what went

wrong.

> Further, there is no evidence that the new {lo} as explained does

> anything to help this situation.

No evidence, I suppose, beyond example sentences, definitions that some

people manage to understand, and the elucidation of a rather common

mental process related to it.

> Depending on the context, "any doctor" would seem to be either {lo

> mikce} (old sense) or {ro mikce}.

Are you claiming that the old lo was more suited to "any" than the new one?

>And, of course there is not problem in asking for a doctor with the old sense of of {lo} (at worst you have to remember that imperatives are intensional contexts, but that is obvious)..

>

When I mean "lo" I mean it without tu'a, le ka, or any of the other

contraptions you've suggested. Are you now claiming that a bare {djica

lo mikce} will suffice???? By now I'm afraid the context is lost, so:

You asked what problems are being fixed, and I responded with "the

ability to use lo mikce for "any doctor" without cmavo torture and a

fortnight of heated discussion each time. "

> The one problem that I have ever seen that might occasionally cause trouble for us learners is theat of negation-transparency and solutions to that can be made up within old {lo} at minimal cost. And even without a change, the solution is just to get used to thinking about what negations do and speaking accordingly.

>

>

>

>*****Several mentioned in passing here and others in the xorxes notes. These may not be real counterexamples to new {lo} or they may be proofs that new {lo} is incoherent. At the moment I am torn, but come down of the middle position that, even if new {lo} is coherent, it is an unnecesssary change, a misguided (probably malglico) attempt to solve some problems basically unrelated to {lo}.

>

>

>

Unfortunately, pc, your response was rather somewhat short on concrete

examples, so as I review my post, I'm not sure it will actually

contribute to any convergence of understanding. To continue fruitfully

we'll need example sentences where the new lo leads to contradictions,

or where a meaning of the old lo is orphaned and can no longer be expressed.

>xod wrote:

>John E Clifford wrote:

>

>

>>But asking for a doctor is just the sort of opaque reference where — as you say — the real existence is irrelevant. Reporting what a doctor did when there are no doctors is another matter and one where existence is relevant.

>>

>>

>>

>>

>

>*It is interesting that existence has historically plagued lo, when we

>see it is really more of a factor for le.

>

>

>

>>As for {mi djica lo ractu}, I can think of several ways whereby this would be a meaningful way to say I want a rabbit. Unfortunately, none of them seem to be the way that {lo} is presently (incoherently) explained.

>>

>>

>>

>

>**Perhaps then you should try working backwards from the premise: the

>proposed lo is intended to work in that sentence as "a rabbit".

>

>

>

>>And, of course, almost all of them leave problems problems in other places (contrasts with {le} in some contexts, permissible external references, and the like) which are avoided by the old Lojban locution.

>>

>>

>>

>

>***Can you offer concrete examples of these problems?

>

>

>

>

>>Can you — xorxes having so far not — provide a clear case of properly used old {lo} which causes a problem which new {lo} solves?

>>

>>

>>

>>

>

>****An end to erroneous idea that lo provides an existence claim and a

>secret "there is a" phrase, and the ability to use lo mikce for "any

>doctor" without cmavo torture and a fortnight of heated discussion each

>time. Paraphrasing Robin: any language in which you cannot ask for a

>doctor is a toy!

>

>

>*****Once again, and if you have concrete counterexample use cases, I'd be

>interested to see them:

>

>

>

>>>There are only 2 actual use cases we must cover.

>>>

>>>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaisanji;

>>>referring to some real objects; the speaker has revealed all of his

>>>criteria; uses lo)

>>>

>>>2. I want the rabbit, a certain rabbit, not any other. (kairsanji;

>>>referring to some real objects, the speaker has not revealed all

>>>criteria but retains some unspoken — he'll let you know if you bring

>>>the wrong rabbit; uses le)

>>>

>>>

>>>

>>>

>

>

>

>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 28 of May, 2004 00:11 GMT

1. How Whorfian! To me it is a matter of expreience: I have met rabbits and myself. I can extract stages from myself (or a rabbit) by analysis though not experience them as such (I've tried for half my life as a student of Buddhism but no go — fortunately the Buddhists I know admit they cqan't either or not for more than a flash). I cannot at all, by synthesis, get Mr. Rabbit from rabbits. So, I have some doubts about stages, none about people or rabbits, and quite a few about Mr. Rabbit, however defined or described.

2. Nice one! In that case the quantifiers are no better than any other device. We need a general solution for this and I never can decide or get an agreement about the best one — or just to pick one (intensional contexts, outer domain quantification, to name the two easiest)

3. I am not sure that I agree that {lo} ever was equivalent to {su'o} and certtainly that fact alone wouldn't justify changing it, unless there were a new, closely related function that needed doing. So far the functions proposed seem either not new or not related. Appeal to Loglan {lo} will have no effect on me (or others who were in that world) than to convince us that your {lo} is inciherent if not contradictory — the status of Loglan {lo} when last I checked.

4: That this is new {lo} is less than clear, BPFK was supposed to clarify and regularize existing forms, not introduce innovations — except to acheve those mentioned tasks. This is new and surely does nothing for either of the set purposes. If you are doing something else, you should announce it loud and clear at the beginning.

5: Yes, fooled by English cleft sentence constructions Lojban creators brook up the single thread of those notions into two places — less drastically in this case than in some others perhaps, but still creating a messy situation. You cannot, for example, officially anaphorize the sumti behind {tu'a} with a coreferential pronoun (you can with a literal one, of couse, but then it means something different. Notice that the third place is also intensional so the {lo tanxe} doesn't create any problem here except that, being in a different intensional context, it cannot be hooked up to the earlier one (I suppose the ideal embedded predicate is {pilno}, which I would use if I wanted to express purpose).

6: but lo ractu does not ractu — no ractu has instances but, at least for now, lo ractu does (How would you say that in Lojban, by the way). I still think you are trying to have things both ways — a constant that does exactly the work of a variable — and very little you have said convinces me otherwise (even leads me to consider it).

Jorge Llambías wrote:

pc:

> F:But the disanalogy is stronger-- one is the product of analysis, the other

> of synthesis, which require very different sorts of things. We know that the

> totality exists and have doubts about the stages in the one case; we jknow

> that the instances exist but have doubts about the totality in the other.

1.I don't have more doubts about the stages than the individual, or

about the kind than the instances. They are all artifacts of language

to me. Maybe that's why we can't seem to understand each other.

> L: But the paradox disappears completely with quantification (one of the

> reasons for its development, in fact). {su'o pavyseljirna na zasti}

2. What about {su'o pavyseljirna cu xanri}?

> M: But why do you think that {lo ractu} old style would be used to say

> something about the class of rabbits or the collective of rabbits or whatever

> it is that you seem to think "rabbits" refers to in English/ {lo} makes a

> lousy tense marker, too, but that is not a reason to replace it; it is a

> reason to get a tense marker that does what is wanted.

3. {lo} was redundant in its function, being equivalent to {su'o},

so it was the logical choice. Also, there's the historical conexion

to Loglan {lo}. Also, the simplest gadri should be the most general

one. {lo} is to gadri as {cu} is to tenses.

> If what you say about the {lo} examples — I have to admit I stopped at {...

> nitcu lo ...}, then the gadri page is in worse shape than even I thought,

> since, by your description, it gives as exemplary cases of {lo} things that

> are not (clearly, generally agreed) cases of {lo} at all.

4:They are examples of the proposed lo, of course. This is the adrees

of the page, in case you want to discuss actual examples:

http://www.lojban.org/tiki/BPFK+Section%3A+gadri

> Byt the way, somehwere earlier you called {tu'a lo ractu} vague. It is, of

> course, since it omits information about what I want/need/etc. for

5:Both {djica} and {nitcu} have an x3 for "for", so the vagueness

of tu'a is of a different sort. For example:

mi nitcu tu'a lo tanxe lo nu mi punji lo cukta ty(?)

I need (something about) a box so that I put some books in it(?)

The "something" is probably "having it". I'm not sure if the {ty}

is correct here. Can you refer to a quantified something that's

inside a different abstraction?

> dreamed about it and so on. But this is marked less vague than {lo ractu} in

> your usage, which also omits all that information (and does not even indicate

> that it may be significant) but plunks us down with something that is not

> even a rabbit (or is only in an indirect sort of way).

6:lo ractu does, of course, ractu. {mi nitcu lo tanxe lo nu mi punji

lo cukta ty} is simply "I need a box to put books in it", no more vague

than that.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 01:26 GMT

1. Assuming the differences you talk about are those for {le} and {lo}, this looks approximately right: for {le} the existence is implicit as a [recondition for using the form, for {lo} existence is actually part of what is said.

2. Fronting is move a referential expression from its proper place in the utterance to the prenex, leaving some appopriate relic behind: {ty broda ro da} to {roda zo'u ty broda da} as a dumb example. The problem cases are one like {mi djica (tu'a) lo tanxe} to {su'o tanxe goi ko'a zo'u mi djica (tu'a) ko'a} This move can't happen with {tu'a} which explicitly blocks it. There seems to be nothing to block it in the case without the {tu'a}.

3: I am not sure I can be clearer without developing the theories (and that is not conducive to clarity either), however the various cases have been discussed in earlier exchanges, going back a decade or more, and so should be in the archives in a gadri thread.

4: Well, it is certainly possibly true here and now that I want a unicorn, but it is certain false of some unicorn I want it. There are no unicorns (and thus, as xorxes says sometimes, no Mr. Unicorn either).

5. Yes, in the sense that {lo} is meant to translate (probably in some of the others as well, but they tend to be in restricted scoping situations — subjunctive and the like).

6: My story about what went wrong tends to be about people misusing {lo} in its normal case and trying to make it do work it very remote cases. And then when they got into trouble they decided that there was something wrong with {lo} because surely they could not make mistakes in Lojban! So they started muckjing with it, resulting in the present impenetrable state. Happily, in most cases — most of the cases of proper use of (old) {lo} — the two uses coincide and the differences are merely in the explanation of what is going on. The remote cases (and a few normal ones) work differently and those are what need soting out. Why is {lo mikce} for "a doctor"controversial with the new {lo}, that is (usually, I need to see a case to be sure) a central usage andm ought to agree with the old.

7:Including those, or what purport to be them. The "elucidations" are garbled at least, the examples are questionable — even in terms of the explanations given, and the mental processes are not ones that I can find in me or in books on the subject.

8: I suppose so, since the old {lo} clearly worked for some cases of "any" and the new one does not clearly work for anything, though seems to be just like old {lo} in the central case (which include the relevant uses of "any").

9: I seem to have missed the discussion. As for the cmavo, well, sometimes they have to be there to say what you want to say. And not always where you would expect them from English. You can say {mi djica lo mikce} and mean to say that you want a doctor and may even be understood to have said it, but the trip to that understanding (other than the patronizing "Well, he is new at this and can't be expected to get it exactly right") is much more complicated than with {tu'a}. In the {lo mikce} for "any doctor" case, I would need to see the whole context; translating "any doctor" alone is just not possible, since it can mean a variety of things depending on what role it is playing in what sentence.

xod wrote:

John E Clifford wrote:

> * Why more {le} than {lo}? Both claim existence, though in different

> way. But then I don't see it as a problem for {lo} even.

1.In kairsanji, existence is implied because the thing existed at least

when the speaker made its acquaintance. In selkaisanji, existence is not

implied at all; the speaker is only discussing characteristics.

> ** that is my premise (or rather the premise of earlier Lojban): {lo}

> is equivalent to a particular quantifier (except perhaps

> distributionally). And it is that in all contexts. In event clauses

> (at least) its range is limited to the world of the mentioned event,

> which may be different from the existents outside in this world. This

> fact keeps us from carrying that quantifier outward ({tu'a} is merely

> a shorthand for an abstract description and so has the same effect).

> The problem with {mi djica lo ractu} is that it looks like that

> reference can be fronted as though it certainly referred to something

> in this world,

2.What does "fronted" mean? Please give an example of the problem it poses.

> and aparently it can be with new {lo}, with all manner of strange

> consequences (changing true to false as far as I can see).

      • Not

> probably without developing the various ways of reading things. But

> briefly, (about {le}) if we make {lo ractu} really about a single

> thing, then all predicates to which it can be attached have to contain

> a reference to a manifestation of that thing, which then will

> conmflict with a {le} sumti, which presumably does not refer to

> something that has manifestations. So, every predicate becomes ambiguous.

3:I don't follow this at all.

> And of course (about distribution), even if {mi djica lo pavyseljirna}

> is true — as it well may be (despite xorxes saying thing which seem

> to require it to be false). {lo pavyseljirna goi ko'a z'u mi djica

> ko'a} is clearly false, since there are no unicorns.

4:Why would this be more false than with the substitution of ko'a with its

value?

> ****How are these problems? The first is just part of what "a"

> basically means in English and has carried over virtually unchanged to

> Lojban.

5.I was talking about the existence claim. Does English "a" have one?

> The second is (if I understand the case) nothing to do with {lo} at

> all and so can't be counted against it.

6.If you read the description in the Book, you'll find it comes very close

to what Jorge is proposing, and that at one time, the idea of using lo

for "a doctor" would not have been controversial. I don't know what went

wrong.

> Further, there is no evidence that the new {lo} as explained does

> anything to help this situation.

7:No evidence, I suppose, beyond example sentences, definitions that some

people manage to understand, and the elucidation of a rather common

mental process related to it.

> Depending on the context, "any doctor" would seem to be either {lo

> mikce} (old sense) or {ro mikce}.

8:Are you claiming that the old lo was more suited to "any" than the new one?

>And, of course there is not problem in asking for a doctor with the old sense of of {lo} (at worst you have to remember that imperatives are intensional contexts, but that is obvious)..

>

9:When I mean "lo" I mean it without tu'a, le ka, or any of the other

contraptions you've suggested. Are you now claiming that a bare {djica

lo mikce} will suffice???? By now I'm afraid the context is lost, so:

You asked what problems are being fixed, and I responded with "the

ability to use lo mikce for "any doctor" without cmavo torture and a

fortnight of heated discussion each time. "

10: Since my comments have mainly been about the metaphysics of {lo ractu} and Mr. Rabbit, examples seem irrelevant. However, the cases at issue have been discussed ad nauseam on several different occasions over the last few decades. They are all there in the archives and if you want to join the discuswsion you need to get up to speed again (you have been there in the past).

> The one problem that I have ever seen that might occasionally cause trouble for us learners is theat of negation-transparency and solutions to that can be made up within old {lo} at minimal cost. And even without a change, the solution is just to get used to thinking about what negations do and speaking accordingly.

>

>

>

>*****Several mentioned in passing here and others in the xorxes notes. These may not be real counterexamples to new {lo} or they may be proofs that new {lo} is incoherent. At the moment I am torn, but come down of the middle position that, even if new {lo} is coherent, it is an unnecesssary change, a misguided (probably malglico) attempt to solve some problems basically unrelated to {lo}.

>

>

>

10:Unfortunately, pc, your response was rather somewhat short on concrete

examples, so as I review my post, I'm not sure it will actually

contribute to any convergence of understanding. To continue fruitfully

we'll need example sentences where the new lo leads to contradictions,

or where a meaning of the old lo is orphaned and can no longer be expressed.

>xod wrote:

>John E Clifford wrote:

>

>

>>But asking for a doctor is just the sort of opaque reference where — as you say — the real existence is irrelevant. Reporting what a doctor did when there are no doctors is another matter and one where existence is relevant.

>>

>>

>>

>>

>

>*It is interesting that existence has historically plagued lo, when we

>see it is really more of a factor for le.

>

>

>

>>As for {mi djica lo ractu}, I can think of several ways whereby this would be a meaningful way to say I want a rabbit. Unfortunately, none of them seem to be the way that {lo} is presently (incoherently) explained.

>>

>>

>>

>

>**Perhaps then you should try working backwards from the premise: the

>proposed lo is intended to work in that sentence as "a rabbit".

>

>

>

>>And, of course, almost all of them leave problems problems in other places (contrasts with {le} in some contexts, permissible external references, and the like) which are avoided by the old Lojban locution.

>>

>>

>>

>

>***Can you offer concrete examples of these problems?

>

>

>

>

>>Can you — xorxes having so far not — provide a clear case of properly used old {lo} which causes a problem which new {lo} solves?

>>

>>

>>

>>

>

>****An end to erroneous idea that lo provides an existence claim and a

>secret "there is a" phrase, and the ability to use lo mikce for "any

>doctor" without cmavo torture and a fortnight of heated discussion each

>time. Paraphrasing Robin: any language in which you cannot ask for a

>doctor is a toy!

>

>

>*****Once again, and if you have concrete counterexample use cases, I'd be

>interested to see them:

>

>

>

>>>There are only 2 actual use cases we must cover.

>>>

>>>1. I want a rabbit, any rabbit, it doesn't matter which one. (selkaisanji;

>>>referring to some real objects; the speaker has revealed all of his

>>>criteria; uses lo)

>>>

>>>2. I want the rabbit, a certain rabbit, not any other. (kairsanji;

>>>referring to some real objects, the speaker has not revealed all

>>>criteria but retains some unspoken — he'll let you know if you bring

>>>the wrong rabbit; uses le)

>>>

>>>

>>>

>>>

>

>

>

>

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 28 of May, 2004 01:26 GMT


> 2. Nice one! In that case the quantifiers are no better than any other

> device. We need a general solution for this and I never can decide or get an

> agreement about the best one — or just to pick one (intensional contexts,

> outer domain quantification, to name the two easiest)

(Re: lo pavyseljirna e la meripapnz cu xanri}

Whatever the solution is, it will apply to Mr Unicorn and to Mary Poppins

in the same way.

> 3. I am not sure that I agree that {lo} ever was equivalent to {su'o} and

But then, what is the old lo you keep talking about if not the one

equivalent to su'o???

> certtainly that fact alone wouldn't justify changing it, unless there were a

> new, closely related function that needed doing. So far the functions

> proposed seem either not new or not related. Appeal to Loglan {lo} will have

> no effect on me (or others who were in that world) than to convince us that

> your {lo} is inciherent if not contradictory — the status of Loglan {lo}

> when last I checked.

OK. In any case, you are fully justified in opposing the new {lo} on

the grounds that it goes against Lojban traditional understanding of it.

I don't really want to argue the political side of it, just the technical

one.

> 4: That this is new {lo} is less than clear, BPFK was supposed to clarify and

> regularize existing forms, not introduce innovations — except to acheve

> those mentioned tasks. This is new and surely does nothing for either of the

> set purposes. If you are doing something else, you should announce it loud

> and clear at the beginning.

There was a large enough consensus that the gadri system as it was needed

fixing. You are more than welcome to propose another way of doing it, but

it has to be more than a sketch, you have to show how to deal with all

the tough cases. It's not enough to claim it can be done, you have to show

how it's done, with concrete examples.

> 5: Yes, fooled by English cleft sentence constructions Lojban creators brook

> up the single thread of those notions into two places — less drastically in

> this case than in some others perhaps, but still creating a messy situation.

(Re: third place of nitcu/djica)

When we use the language, we have to deal with the messy situation,

so it is not enough to say that it could have been done better, we

have to show how to deal with it as it is.

> You cannot, for example, officially anaphorize the sumti behind {tu'a} with a

> coreferential pronoun (you can with a literal one, of couse, but then it

> means something different. Notice that the third place is also intensional

> so the {lo tanxe} doesn't create any problem here except that, being in a

> different intensional context, it cannot be hooked up to the earlier one (I

> suppose the ideal embedded predicate is {pilno}, which I would use if I

> wanted to express purpose).

So what do you propose we do about it?

> 6: but lo ractu does not ractu — no ractu has instances but, at least for

> now, lo ractu does (How would you say that in Lojban, by the way).

But rabbits do rabbit, and that's what {lo ractu cu ractu} means.

As for the rest, I'm not sure, but wouldn't every ractu be an instance

(the only instance) of itself? I'm just pondering aloud here, I know

what your answer will be.

In any case, in a more general case, instances of one kind can in

turn be a kind with its own instances: for example rabbits as a kind

are an instance of animals. (Each rabbit is also an instance, of course.)

I think {mupli} could be used for "x1 is an instance of x2", modifying

its place structure a bit, and {klesi} for subkinds, so:

ta mupli lo ractu

That's an instance of rabbits.

lo ractu cu klesi lo danlu

Rabbits are a kind of animal.

> I still

> think you are trying to have things both ways — a constant that does exactly

> the work of a variable — and very little you have said convinces me

> otherwise (even leads me to consider it).

As long as it does the work, what's the problem?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 02:17 GMT

pc:

> 4: Well, it is certainly possibly true here and now that I want a unicorn,

> but it is certain false of some unicorn I want it. There are no unicorns

> (and thus, as xorxes says sometimes, no Mr. Unicorn either).

I make a distinction between the predicate {zasti} and the "there is"

of existential quantification. I have no problem with:

su'o da naku zasti

There are things that don't exist (in this world).

I wouldn't say that there is no Mr Unicorn. I'd say that Mr Unicorn

is such that he does not exist in this world (and the same thing I

would say of Mary Poppins, for example.)

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 15:31 GMT

7. Probably, though at least (old) {lo} might take a slight change, playing off its relation to {su'o}

8: Well, it is eqquivalent in most cases, but I have proposed a couple of times that there be differnces in a few contextx. In particular, in opaque contexts I would use {su'o} to inidcate that the intended item was clearly in the real world while {loaaaaaaaaaaaa] left that question open. It is just a proposal but I don't want to cut it off by some admission I make in a different context.

9: I agreee and it is precisely the technical one that I find so deplorable (admittedly from the historical point of view initially but then from a purely tchnical one as well).

10: Back atcha! Your claims that new {lo} solves some of these problems are at best tendentious when not clearly false. I would solve these "problems" at least pro tem simply by adding new gadri (or other devices) with the intended functions (vocab out of xCC, I suppose) and then see if 1) they get any use at all and 2) see if ways can be found to do without them using existing elements. I would not do it by redefining a stable form.

I am not convinced that the gadri system needs more reworking than that (and a change on the meaning of internal and external quantifiers, perhaps).

11: Done, though it involves just not using {nitcu3} with {nitcu2} {mi nitcu tu'a lo tanxe poi mi punji lo cukta ke'a}

12:That makes two proposals — not mutually exclusive. I suppose I could come up with more. How many do you need?

13: Your {lo ractu} is a constant, that is it has a single referent, the same in all contexts (so you say), but in different contexts different rabbits are used to make the resulting sentence true. There is no one rabbit that makes all true {lo ractu} sentences true. lo ractu is no more a rabbit than John is a John-stage. It is your saying things like this that convince me that you are constantly confusing (or shifting back and forth between) an abstractum and the underlying concreta.

14: It depends, but for the general case you suggest, the answer it clearly "no" — instance are at a different level from what they are instances of.

15: Yes, and that is exactly the classic tree or hierarchy of abstracta, with all the usual rules. So there clearly are both and abstract something and a concrete one here and the question is only whether they are distinct or confused. If distinct, then either lo ractu is not a ractu or else {lo ractu} is not a constant.

16: Fine, provided that the 2nd place can embrace kinds or whatever as well as properties (assuming that there is a significant difference, which I think there is, though I would be hard-pressed to spell it out).

17: These work fine with {lo ractu} as a kind, but fail when we look (as we usually have to do) for an instance that makes it true. That Mr. Rabbit is eating grass here is true because a rabbit instance is eating grass here. That Mr. Rabbit is a class is not true because some rabbit is. Which one is correct?

18: The point is that when we get down to careful usage — which is what Lojban is about, basically — it does not work. See examples just above.

Jorge Llambías wrote:


> 2. Nice one! In that case the quantifiers are no better than any other

> device. We need a general solution for this and I never can decide or get an

> agreement about the best one — or just to pick one (intensional contexts,

> outer domain quantification, to name the two easiest)

7. Re: lo pavyseljirna e la meripapnz cu xanri}

Whatever the solution is, it will apply to Mr Unicorn and to Mary Poppins

in the same way.

> 3. I am not sure that I agree that {lo} ever was equivalent to {su'o} and

8: But then, what is the old lo you keep talking about if not the one

equivalent to su'o???

> certtainly that fact alone wouldn't justify changing it, unless there were a

> new, closely related function that needed doing. So far the functions

> proposed seem either not new or not related. Appeal to Loglan {lo} will have

> no effect on me (or others who were in that world) than to convince us that

> your {lo} is inciherent if not contradictory — the status of Loglan {lo}

> when last I checked.

9: OK. In any case, you are fully justified in opposing the new {lo} on

the grounds that it goes against Lojban traditional understanding of it.

I don't really want to argue the political side of it, just the technical

one.

> 4: That this is new {lo} is less than clear, BPFK was supposed to clarify and

> regularize existing forms, not introduce innovations — except to acheve

> those mentioned tasks. This is new and surely does nothing for either of the

> set purposes. If you are doing something else, you should announce it loud

> and clear at the beginning.

10:There was a large enough consensus that the gadri system as it was needed

fixing. You are more than welcome to propose another way of doing it, but

it has to be more than a sketch, you have to show how to deal with all

the tough cases. It's not enough to claim it can be done, you have to show

how it's done, with concrete examples.

> 5: Yes, fooled by English cleft sentence constructions Lojban creators brook

> up the single thread of those notions into two places — less drastically in

> this case than in some others perhaps, but still creating a messy situation.

11:(Re: third place of nitcu/djica)

When we use the language, we have to deal with the messy situation,

so it is not enough to say that it could have been done better, we

have to show how to deal with it as it is.

> You cannot, for example, officially anaphorize the sumti behind {tu'a} with a

> coreferential pronoun (you can with a literal one, of couse, but then it

> means something different. Notice that the third place is also intensional

> so the {lo tanxe} doesn't create any problem here except that, being in a

> different intensional context, it cannot be hooked up to the earlier one (I

> suppose the ideal embedded predicate is {pilno}, which I would use if I

> wanted to express purpose).

12: So what do you propose we do about it?

> 6: but lo ractu does not ractu — no ractu has instances but, at least for

> now, lo ractu does (How would you say that in Lojban, by the way).

13:But rabbits do rabbit, and that's what {lo ractu cu ractu} means.

14: As for the rest, I'm not sure, but wouldn't every ractu be an instance

(the only instance) of itself? I'm just pondering aloud here, I know

what your answer will be.

15: In any case, in a more general case, instances of one kind can in

turn be a kind with its own instances: for example rabbits as a kind

are an instance of animals. (Each rabbit is also an instance, of course.)

16:I think {mupli} could be used for "x1 is an instance of x2", modifying

its place structure a bit, and {klesi} for subkinds, so:

17:ta mupli lo ractu

That's an instance of rabbits.

lo ractu cu klesi lo danlu

Rabbits are a kind of animal.

> I still

> think you are trying to have things both ways — a constant that does exactly

> the work of a variable — and very little you have said convinces me

> otherwise (even leads me to consider it).

18: As long as it does the work, what's the problem?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 15:31 GMT

19: OK. But you need to make it clear that you are allowing for quantification over non-existents. That is a perfectly good way to go and, once started, seems to require that you go all the way. We probably then need a different set of quantifiers for the existents, since it is often important to know that something really exists. The old rules for quantifiers (for the zastis) are then applied only to this restricted set and the unbounded set covers everything and always applies. I was attributing to you a less drastic change, taking {lo zasti} to apply to a kind, an abstract entity, which by convention would therefore exist, and then (mistakenly) saying that it did not exist since it had no instances. And indeed that is closer to what you now claim, since Mr.Unicorn is said not to exist in this world but to be nonetheless. I'm not sure what that does to the problem inference, but I think it still makes the start {mi djica lo pavyseljirna} false, like the fronted version.

However, this gets tangled with the ambiguity of {lo pavyseljirna} so I am not sure.

I hope, by the way, that the shift to quantifying over the outer domain is made for carefully considered reason, not just to save a few embarassing cases.

Jorge Llambías wrote:

pc:

> 4: Well, it is certainly possibly true here and now that I want a unicorn,

> but it is certain false of some unicorn I want it. There are no unicorns

> (and thus, as xorxes says sometimes, no Mr. Unicorn either).

19:I make a distinction between the predicate {zasti} and the "there is"

of existential quantification. I have no problem with:

su'o da naku zasti

There are things that don't exist (in this world).

I wouldn't say that there is no Mr Unicorn. I'd say that Mr Unicorn

is such that he does not exist in this world (and the same thing I

would say of Mary Poppins, for example.)

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 15:31 GMT

pc:

> [[before:%20I%20am%20not%20sure%20that%20I%20agree%20that%20%7Blo%7D%20ever%20was%20equivalent%20to%20%7Bsu'o%7D|before: I am not sure that I agree that {lo} ever was equivalent to {su'o}]]

> 8: Well, it is eqquivalent in most cases, but I have proposed a couple of

> times that there be differnces in a few contextx. In particular, in opaque

> contexts I would use {su'o} to inidcate that the intended item was clearly in

> the real world while {loaaaaaaaaaaaa] left that question open. It is just a

> proposal but I don't want to cut it off by some admission I make in a

> different context.

That seems to result in something almost indistinguishable from my

proposal, leaving aside the meta-talk.

I removed the nitcu and djica examples from the page. Even though

I will keep using them in the sense "person x1 wants object x2" and

"person x1 needs object x2", the fact is that at least for djica

the gi'uste seems to restrict it to "perxon x1 wants event x2", so

I won't push an example that contradicts the gi'uste. (Also, Robin

wanted less examples from L* P**** P****, so that's an additional

reason to remove them.)

Of the remaining examples in that page, would you say that any

of them conflicts with your understanding of the old {lo}?

How would you improve the wording of the definition of {lo}

to make it coherent?

> 13: Your {lo ractu} is a constant, that is it has a single referent, the same

> in all contexts (so you say), but in different contexts different rabbits are

> used to make the resulting sentence true.

Right, just as in different contexts different John-stages are used

to make the resulting sentence true. (The John-stage that makes true

"John goes to the market on Saturday" is not the John-stage that makes

true "John stayed at home on Sunday".)

> There is no one rabbit that makes

> all true {lo ractu} sentences true.

No single instance of rabbits, right.

>lo ractu is no more a rabbit than John is

> a John-stage.

Using {le ca me la djan} for now-John, and {le puza me la djan} for

a-while-ago-John, then I would say:

le ca me la djan cu me la djan

Now-John is John.

le puza me la djan cu me la djan

A-while-ago-John is John.

la djan cu me la djan

John is John.

So yes, the same identificatory predicate that is satisfied by the

stages is satisfied by the individual, and the same I would say

for kinds and instances.

(I would like to be able to talk of {la ca djan} and {la puza djan},

but the current grammar forbids it.)

> That Mr. Rabbit is

> eating grass here is true because a rabbit instance is eating grass here.

> That Mr. Rabbit is a class is not true because some rabbit is. Which one is

> correct?

Both are correct, because in general the truth of {lo broda cu brode}

does not hang on the truth of {su'o mupli be lo broda cu brode}. For

some predicates brode, it just happens that the second entails the first,

but that's due to the semantics of the predicate, not due to any logical

necessity.

Anyway, if you can check the list of examples and tell me which ones

look wrong to you (and how you would correct them) that would be

helpful.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 15:32 GMT

pc:

> 19: OK. But you need to make it clear that you are allowing for

> quantification over non-existents.

That depends on the context, doesn't it? Surely sometimes you

quantify over non-existents when speaking in English and other

times you don't. If context does not make it clear that you

are restricting it to existents, you can always say

{ro da poi zasti zi'e poi ...}, {su'o da poi zasti zi'e poi ...}.

> That is a perfectly good way to go and,

> once started, seems to require that you go all the way. We probably then

> need a different set of quantifiers for the existents, since it is often

> important to know that something really exists.

When it is not obvious what you mean, you can explicitly restrict

with {poi zasti}. In a similar way, sometimes we use {no da} when

we mean {no da poi prenu}, and so on. If you want to buy more

precision you have to pay with more verbosity.

> I hope, by the way, that the shift to quantifying over the outer domain is

> made for carefully considered reason, not just to save a few embarassing

> cases.

Shift?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 17:29 GMT

1. To be sure, since your proposal was at least conservative enough to preserve the core cases of old {lo}. The differences would come only in the new uses, which I simply would not take to be cases of {lo} (I have fleshed out these suggestion a bit on a comment to the gadri page.)

2. What?

3: I blushingly confess that I have not studied the examples, but on a glance I saw at least one ("A young person respect an old person" or so) that I agree should be {lo} (given that {ei} generates an opaque context, or rather one tha delimits any quantification) but that others might want to put (with some justification) into the species class.

4: I think it did that in the note, something like "a (or several) thing(s) in the current domain that actually meet(s) the given property." which probably needs some tidying. I think your stuff about quantities can stand (thouggh, as I note, some of the more remote cases don't make sense).

5: Well, I still don't accept the analogy for the reasons rehearsed before, so this is not a convinving case. And it certainly does not override the problems with taking {lo} for Mr. Rabbit when his properties qua Mr. Rabbit are at issue.

6: Since I don't know what {me} is meaning this week, I can't be sure what to make of these sentences. Assuming that the wordlist is approximately right, then I think that saying John-now is specific to John will work only with the tag "in respect to time-stages" or some such. I suppose the corresponding case you want is {leva ractu me lo ractu} or so, which again may work with some tag. But that is not the case that is interesting (even if it does work with a nonvacuous respect (it looks a lot like "this rabbit is a rabbit" with some fluff thrown in). In any case, the identificatory predicates will not be same — only the {me} carries over, not the aspect.

7: Logic hjabits die hard, so I like the already permissible (I think) {la djan xi ca}. As I write this I notice that {xi} seems only defined for numerals and the like, so some tampering is required here as well (or with {la}, though that seems to me harder).

8: I would want to go back to the old system where {lo broda cu brode} does entail (indeed is equivalent to in this context?) {su'o mupli be le broda cu brode}. Of course, your replacing the proeprty in {mupli2} with a whatever pretty much prejudices the issue. But notice at what price simplification has been bought here: every place of every predicate has to be marked for whether {lo broda} in that place refers to some broda or to Mr. Broda. And yet there will be cases that are hard to pin down {lo broda cu zasti} is ambiguous; it doesn't make a difference since in your strange halfbreed Mr. Broda they are extensionally equivalent. And, of course, getting out of extensional case separates them: what am I thinking about when {mi pensi lo ractu} is true?

Jorge Llambías wrote:

pc:

> [[before:%20I%20am%20not%20sure%20that%20I%20agree%20that%20%7Blo%7D%20ever%20was%20equivalent%20to%20%7Bsu'o%7D|before: I am not sure that I agree that {lo} ever was equivalent to {su'o}]]

> 8: Well, it is eqquivalent in most cases, but I have proposed a couple of

> times that there be differnces in a few contextx. In particular, in opaque

> contexts I would use {su'o} to inidcate that the intended item was clearly in

> the real world while {loaaaaaaaaaaaa] left that question open. It is just a

> proposal but I don't want to cut it off by some admission I make in a

> different context.

1.That seems to result in something almost indistinguishable from my

proposal, leaving aside the meta-talk.

I removed the nitcu and djica examples from the page. Even though

I will keep using them in the sense "person x1 wants object x2" and

"person x1 needs object x2", the fact is that at least for djica

the gi'uste seems to restrict it to "perxon x1 wants event x2", so

I won't push an example that contradicts the gi'uste. (Also, Robin

wanted less examples from 2.L* P**** P****, so that's an additional

reason to remove them.)

3:Of the remaining examples in that page, would you say that any

of them conflicts with your understanding of the old {lo}?

4:How would you improve the wording of the definition of {lo}

to make it coherent?

> 13: Your {lo ractu} is a constant, that is it has a single referent, the same

> in all contexts (so you say), but in different contexts different rabbits are

> used to make the resulting sentence true.

5:Right, just as in different contexts different John-stages are used

to make the resulting sentence true. (The John-stage that makes true

"John goes to the market on Saturday" is not the John-stage that makes

true "John stayed at home on Sunday".)

> There is no one rabbit that makes

> all true {lo ractu} sentences true.

No single instance of rabbits, right.

>lo ractu is no more a rabbit than John is

> a John-stage.

Using {le ca me la djan} for now-John, and {le puza me la djan} for

a-while-ago-John, then I would say:

6:le ca me la djan cu me la djan

Now-John is John.

le puza me la djan cu me la djan

A-while-ago-John is John.

la djan cu me la djan

John is John.

So yes, the same identificatory predicate that is satisfied by the

stages is satisfied by the individual, and the same I would say

for kinds and instances.

7:(I would like to be able to talk of {la ca djan} and {la puza djan},

but the current grammar forbids it.)

8: That Mr. Rabbit is

> eating grass here is true because a rabbit instance is eating grass here.

> That Mr. Rabbit is a class is not true because some rabbit is. Which one is

> correct?

Both are correct, because in general the truth of {lo broda cu brode}

does not hang on the truth of {su'o mupli be lo broda cu brode}. For

some predicates brode, it just happens that the second entails the first,

but that's due to the semantics of the predicate, not due to any logical

necessity.

Anyway, if you can check the list of examples and tell me which ones

look wrong to you (and how you would correct them) that would be

helpful.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 17:51 GMT

1. I don't think so. Once you do it you seem to be committing the language. To be sure, you could contextualize (put the scope in an opaque clause) and go on from there. This is one standard work-around logicians use for English, which is very sloppy about this. Lojban ought to do better, not leaving it to context. And just how would those contexts be described? What separates {la crlakolmz xabju la lndn} from {la tonibler xabju la lndn} so that we can sort out what is real from what is not. (By the way, I would kee the current quantifiers for existents and add the outer domain one on, just as a practical matter.)

2: Or with a somewhat longer cmavo list, either more quantifiers or perhaps more "trenses" "in story" relevantly in this cse. And, of course, all of these (like everything else in Lojban) are optional in most speech situations. The point is to have the tools when more precision is needed.

3: Shift! The present quantifiers are explicitly real-world (and, indeed, I would have said — if asked — that you were one of the people who jumped on me for suggesting otherwise. Since, in the outer domain, neither {ro} nore {su'o} goes unfufilled, many questions disappear — including, obviously, the one about how many whojis there are for internal quantification. Hooray!)

Jorge Llambías wrote:

pc:

> 19: OK. But you need to make it clear that you are allowing for

> quantification over non-existents.

1. That depends on the context, doesn't it? Surely sometimes you

quantify over non-existents when speaking in English and other

times you don't. If context does not make it clear that you

are restricting it to existents, you can always say

{ro da poi zasti zi'e poi ...}, {su'o da poi zasti zi'e poi ...}.

> That is a perfectly good way to go and,

> once started, seems to require that you go all the way. We probably then

> need a different set of quantifiers for the existents, since it is often

> important to know that something really exists.

2:When it is not obvious what you mean, you can explicitly restrict

with {poi zasti}. In a similar way, sometimes we use {no da} when

we mean {no da poi prenu}, and so on. If you want to buy more

precision you have to pay with more verbosity.

> I hope, by the way, that the shift to quantifying over the outer domain is

> made for carefully considered reason, not just to save a few embarassing

> cases.

3:Shift?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 20:01 GMT


> 2. What?

Le Petit Prince

> 7: Logic hjabits die hard, so I like the already permissible (I think) {la

> djan xi ca}. As I write this I notice that {xi} seems only defined for

> numerals and the like, so some tampering is required here as well (or with

> {la}, though that seems to me harder).

The easiest path is to blend CMENE with BRIVLA. No point in them

having different grammars.

> But notice at what price simplification

> has been bought here: every place of every predicate has to be marked for

> whether {lo broda} in that place refers to some broda or to Mr. Broda.

{lo broda} ALWAYS refers to Mr Broda. No marking is required.

>From {ko'a viska lo broda} we can deduce, based on our

understanding of viska, that {ko'a viska su'o broda}.

But {lo broda} itself is a visible thing. We can say, for example,

pointing to Mr Rabbit, present as a single instance in

front of us:

mi pu viska su'o ta

I saw one of those before.

That doesn't mean necessarily that I saw the same instance.

{ta} in this case refers to Mr Rabbit, just as "those" in English.

> And

> yet there will be cases that are hard to pin down {lo broda cu zasti} is

> ambiguous; it doesn't make a difference since in your strange halfbreed Mr.

> Broda they are extensionally equivalent. And, of course, getting out of

> extensional case separates them: what am I thinking about when {mi pensi lo

> ractu} is true?

You'd be thinking about rabbits, what else?

(I'm copying here some of your comments from the page.

I would prefer that you use the 'discuss' forum rather than the

'comments', because it makes it much easier to reply. Besides, I

don't find out that you have made a comment until I visit the page

or you tell me, whereas the 'discuss' forum I get in my post box.)

pc:

>on {lo'e} and {le'e}.These do NOT take external quantifiers; internal

>quantifiers are about the size of group which is being typed or

>stereotyped (and so is rarely used).

I will write it as you suggest unless there is opposition from others.

The grammar still allows outer quantifiers, but I'm not particularly

interested in assigning them weird meanings.

>If we want to talk about more than one typical whatever, the

>appropriate form (even if it is only one but is not being used for

>typing) is to used the (as yet unlexed) brivla for "typuical"

>and "stereotypical".

{fadni} can be used for "typical". I don't know about "stereotypical".

>Along the lines of these, a modern language surely needs corresponding

>forms for "the average," probably one for each measure of central

>tendency> Of these, the mean does not support generalization but both

>median and mode do, so this fact should be made visible somehow (maybe,

>sticking to the rules, {xo'e} for mean and have the others look more

>like normaler gadri: {xa'a} and {xa'e} say).

There's a page for this kind of thing if you are really interested:

http://www.lojban.org/tiki/Currently+proposed+experimental+cmavo?

I'm curious as to what kind of thing you could say about

the median flower, or the modal rabbit, or the mean shoe.

I just can't think of a place I would use one of those.

>I am unsure about the following additions. Trying to deal with

>them without much care is what has complicated {lo} out of

>coherence. On the other hand, I suspect tha, once we master

>subjunctives in Lojban, the problem cases will be dealt with.

>Until then (and maybe even after then as a shortcut — like {lo'e}

>and {le'e} (although admittedly we cannot now say what is

>abbreviated in these case)):

>a gadri, {xo'o} say, which produces the name of the species/kind/...

>of things that satisfy the predicate. Such things always exist, even

>if there are no such critters as satisy the predicate. These can be

>used to make general claims of one sort.

Of what sort? Can these be used in contrast with {lo broda}

to make a distinction, or only in places where {lo broda} cannot

be used?

>And another for the stuff/substance/goo of things of the predicated

>sort. This is useful for another kind of generalization (though less

>common, I think, except for the normally mass nouns).

Examples?

>Once the extension of old {lo} are dealt with separately, old {lo}

>can take on its proper job again, doing what has rather less

>successfully been done by most uses of new {lo}: an unspecified

>member of the current domain that actually has the proberty described.

>As such, it is generalizable **within the range of the current domain**

>but not generally outside.

Can it be the antecedent of a pronoun from outside the current

domain? For example, I could say:

mi nitcu lo nu mi cpacu lo tanxe

i mi pu viska ty bu'u le lamji kumfa

i e'apei mi lebna ty

"I need to get a box.

I saw ONE in the other room.

May I take IT?"

Can this be done with your proposed {lo}, given that the reference

is from outside the domain where {lo tanxe} appears?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 20:01 GMT

pc:

> And just how would those contexts be described? What

> separates {la crlakolmz xabju la lndn} from {la tonibler xabju la lndn} so

> that we can sort out what is real from what is not.

If the names themselves are not enough, you could add something

explicit {la crlakolmz noi se cfika} or maybe {la crlakolmz xabju

la lndn sei cfika}.

> 2: Or with a somewhat longer cmavo list, either more quantifiers or perhaps

> more "trenses" "in story" relevantly in this cse.

I don't see the Lojban community adopting a whole set of new cmavo

at this stage. One or two maybe, but drastic new incorporations, I

don't see it. More likely is the adoption of something already

existing with perhaps a different or more specialized meaning than

what it was planned for. But you can always propose new ones if you

think they are worth it.

> And, of course, all of

> these (like everything else in Lojban) are optional in most speech

> situations. The point is to have the tools when more precision is needed.

Indeed.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Fri 28 of May, 2004 20:03 GMT posts: 14214

Just for the record, one of the current No votes is from PC, who is not a member of the BPFK, and as such should be considered as being for informational purposes only.

-Robin



Posted by Anonymous on Fri 28 of May, 2004 22:15 GMT

One of my problems with Jorge's proposal is that it seems to either

conflate two different meanings into one, or have one meaning that is

ill-defined or hard to formalize. What I want is a general algorithm fo=

r

determining whether or not a sentence that uses XS-lo is true or false.

These examples are taken from an IRC discussion I had with xod.

{mi nitcu lo mikce} — obviously true (in the right circumstances)

Here, I take every {mikce} in the world, assign it to {xy.} and for eac=

h

one asks if {mi nitcu xy.} is true. If and only if there is no {xy.} fo=

r

which {mi nitcu xy.} is false, then {mi nitcu lo mikce} is true. No

problem here.

{lo prenu cu tarci} — obviously false

We are then looping through every {prenu}. There are some (actually all=

)

instances in which {xy. tarci} is false, therefore {lo prenu cu tarci} =

is

false. No problem here.

{lo prenu cu sadjo} — obviously true.

We are again looping through every {prenu}. There are some instances in

which {xy. sadjo} is false, therefore {lo prenu cu sadjo} is false. Whi=

ch

is patently wrong, but fits with how I understand XS-lo.

So. What is it that I'm missing here.

--=20

Arnt Richard Johansen http://arj.nvg.org=

/

P=E5 1300-tallet kom tersen. F=F8r og etter det var det meste bare rot,=

men

s=E5 kom Sch=F6nberg og ordnet opp. Puh. Endelig litt system. S=E5 klar=

te Arne

Nordheim =E5 rote det til igjen. — Under Dusken 08/2=

001



Posted by Anonymous on Fri 28 of May, 2004 22:15 GMT


> {mi nitcu lo mikce} — obviously true (in the right circumstances)

> Here, I take every {mikce} in the world, assign it to {xy.} and for eac=

> h

> one asks if {mi nitcu xy.} is true. If and only if there is no {xy.} fo=

> r

> which {mi nitcu xy.} is false, then {mi nitcu lo mikce} is true. No

> problem here.

But what you are describing is {mi nitcu ro mikce}.

In {mi nitcu lo mikce} you don't examine any doctor. All you

need to know is what a doctor is. Then, knowing what a doctor

is, you ask yourself, is that what I need? If the answer is yes,

then {mi nitcu lo mikce} is true. You don't have to examine any

doctors. As xod says, all the doctors may be dead and you may

still need one.

> {lo prenu cu tarci} — obviously false

> We are then looping through every {prenu}. There are some (actually all=

> )

> instances in which {xy. tarci} is false, therefore {lo prenu cu tarci} =

> is

> false. No problem here.

Again you are describing the procedure to check {ro prenu cu tarci}.

For {lo prenu cu tarci} what you need to know is what a prenu is.

Knowing what a prenu is, do you think it is (at least sometimes)

a tarci? If not, then the sentence is false.

> {lo prenu cu sadjo} — obviously true.

> We are again looping through every {prenu}. There are some instances in

> which {xy. sadjo} is false, therefore {lo prenu cu sadjo} is false. Whi=

> ch

> is patently wrong, but fits with how I understand XS-lo.

Again, you are describing {ro prenu cu sadjo}, which as you conclude

is false.

To check {lo prenu cu sadjo} you start from your knowledge of

what a prenu is. Then you ask yourself, is it (at least sometimes)

a sadjo? If yes then {lo prenu cu (su'oroi) sadjo} is true.

{lo prenu roroi fe'eroroi sadjo} is obviously false. People can be

Saudis, but people are not always and everywhere Saudis.

> So. What is it that I'm missing here.

You seem to be confusing lo with ro.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 28 of May, 2004 23:55 GMT

About the {lo} examples:

General comment: most of these examples are complex; good examples begin, at least, with clear simple cases where there is little chance to miss the point. Such examples are boring, but they provide the information readers are looking for at the start. Later cases can deal with peculiarities. And most of these are peculiar cases (if really cases of {lo} at all): generalities, gnomic utterances, maxims and the like – things that are more or less universal; that is not {lo} home ground (and very likely not its ground at all).

ei lo verba cu mutce fraxu lo makcu prenu

Children should always show great forbearance

toward grown-up people.

This looks like a general maxim or moral rule. It is unlikely that the propounder would think it satisfied if just one child was found who forgave one adult. Using {ro} instead of {lo} is probably closer to the point to be made, especially if we allow implicit exceptions (as morals pretty regularly do) to at least the second {ro}: adult ax murderers are not to be forgiven or are their child victims required to forgive them. In these fuzzy areas (as maxims usually are) the temptation to talk in terms of “some” – which includes “all” and everything in between — is pretty strong but gives the wrong effect, as noted. Trying to do it with classes/species/kinds raises the same problems both ways (not to mention the problem of how to word it sensibly). Of course the propounder might (probably does in typical cases) has some particular youths in mind and so even {le} could be justified on that occasion.

ku'i uinai mi na viska lo lanme pa'o lo tanxe

i ju'ocu'i mi milxe simsa lo makcu prenu

But I, alas, do not see sheep through the walls of

boxes. Perhaps I am a little like the grown-ups.

I am not sure whether {pa’o} works like this, but the {lo}s in the first sentence work out right. A good example (though perhaps for later), since it reminds us that universals in negative contexts are expressed existentially: “any sheep through any box” (is “the walls of” just a flourish? This eems to apply as well to looking through a tubular box lacking both ends. The {lo} in the second sentence is probably about a species (etc.) since it is going on to some property. I would use {la’e} here, but that is only a reasonable start of working out how to talk about species.

ca lo nicte lo cinfo cu kalte lo cidja

At night lions hunt for food.

As usual for generalities, the propounder would not feel he had made his point if only on one night was one lion found to be hunting food – a single eland, say. But of course the universal is not required either, since some lions sometimes take a night off (after a big meal or when they have some fleshy corpse still available. (Lions are actually day hunters at least as much as night, but that is not relevant here.) This really soes seem to be about the species: “Lions are nocturnal food-hunters,” however that works out in the end (pretty much like that or in terms of explicit relation between species or between a species and a property — or maybe all three and more besides.)

lo pa pixra cu se vamji lo ki'o valsi

One picture is worth a thousand words.

Ah, I forgot this aspect of your work with quantifiers. {lo ki’o valsi} looks OK and not noticeably different from {ki’o valsi} – presumably the words could be spelled out in each case, maybe several different ways, indeed. Presumably this is gnomic again so the first {lo} is either universal or about species or perhaps {la’e}.

de'i li 1960 lo pare sovda cu fepni li 42

In 1960 a dozen eggs cost 42 cents.

Same old, same old. It was not just one dozen but just about any dozen there was – implicit exception in force (crested floo-floo birds’ – now extinct – eggs, certified organic, …). Iam inclining more and more to {la’e} here.

lo ctuca cu fendi lo selctu mu lo vo tadni

The teacher will divide the class

into five groups of four students.

Hey, some basic cases, though {le ctuca} makes better sense — this seems to be a particular occasion. So, come to that, {le selctu} or even {lei selctu}. But the {mu lo vo tadni} is nice.

lo bidjylinsi pe lo ze seldri cu se pagbu

ze lo ze bidju e ji'a ci lo pa bidju e lo kucysni

The Rosary of the Seven Sorrows consists of seven groups

of seven beads, with three additional beads and a Crucifix.

This looks like it is about a certain class of things, a particular kind of rosary (and, indeed, if it was about a unique thing {le} would be appropriate. Here there is none of the worry about exceptions that the more gnomic cases call for, so this could be done with {ro}. But I take it to be about the kind, laying out its particularities. In that case, the last three {lo}s are just any-olds; put them together in this way and you get a rosaary of the right sort. The first should be for species or kind and whether this form or some other covers these cases I leave for a while.

o'i mu (lo) xagji sofybakni cu zvati le purdi

Caution! There are five hungry Soviet cows in the garden.

No problems that I can see (finally!) This should come much earlier in the business.

lo sanli darxi bo dakli cu culno lo djacu onai lo canre to lo djacu

cu pukmau ki'u lo nu slilu tolcando toi gi'e bunda li ji'i 270

Standing punching bags are filled with water or sand - water

being preferable because of the wave-motion created - and

weigh about 270lbs.

Species substance substance substance species (but maybe, in all this scope, {lo} would work)

lo pavyseljirna cu ranmi danlu gi'e catlu lo ka ge ce'u xirma

gi lo jirna cu cpana lo sedycra be ce'u

Unicorns are mythical creatures that look like a horse

with a horn coming out of their foreheads.

Species conventional (could be {le} just as well) ok conventional (but I think {le} is a more sensible convention). This looks like a good safish way to talk about species (well, with the appropriate gadri, of course).

bilga lenu jdice lenu roroi pilno lo mokla tirxe

(to zoigy. velar gy. toi) jonai crane (to zoigy.

alveolar gy. toi)

tavla fi le tutra pe le terdi

I’m not sure about the context here, but this looks ok: on each occasion one use some velar (or alveolar). But complex for the point. How is this a problem solved; it seems to be basic {lo} What is the role of the blue expressions?

le cmana lo cidja ba claxu

In the mountains there is no food.

lapoi pelxu ku'o trajynobli

Normal usage – well it is good to see that implicit negation works like ex-lciti (but does it? I hope so).

le dargu pe lo xamgu bangu cu kargu

The road of the good language is costly.

lapoi pelxu ku'o trajynobli

Specific or universal (probably the latter — it seems merely factual)

la jyryr. tolkien. cu te cukta la djine turni (to la'o

gy Lord of the Rings gy toi) .e le so'omoi be lo

xanri munje lisri ca le lampru na'acto

tenguar

Species or set (probably the latter). “the severalth” is nice, though not a clear as it might be; I suppose it is to me “one of several” or just “pretty far along in the set ordered by … (date?)”



Posted by Anonymous on Sat 29 of May, 2004 16:46 GMT

I agree with the opening remark here, though for somewhat different reasons. The problem that arj is dealing with arises from the fact that one (or maybe the only) "meaning" assigned to {lo} is essentially vague. Trying to apply recise tests to discover the truth of a sentence in this case is bound to fail. Looking at the examples on the wiki sheet I found that most of them fluctuated (when applying precise tests) between "all" and "some" but with conditions. Thus, if read as "some," it often turned out that one or two — or any fixed number of — cases were not enought; if read as "all" then any number of exceptions had to be allowed without affecting truth. Nor would any of "may", "most," or such intermediate — and already rather vague — quantifiers work. This is out of the range of quantifers and into operators like "generally," "as a rule" and so on, applied at the argument level rather than the sentential. This seems to be one thing that xorxes means when he says that

{lo} expressions refer to kinds or species or maybe even Mr. The next step (and where the ambiguity comes in from my point of view) is moving from talking about things in a certain unspecified collective way to talking about one thing, the unspecified collect of those things — more or less like taking the typical x as being some paricular (though prehaps unidentified) x. This way of talking does result in an expression that serves much like a constant: it is usually negation transparent — if it is not the case that some unspecified collective of xs are ys, then such a collective is usually not ys (unles there are no xs at all) and, except for discrete properties, is such a collective is a and such a collective is b, then there will be such a collective that is both a and b. These are not valid moves, of course, but they usually work and thus can be applied in practical cases.

BTW, the test for {mi nitcu lo mikce} is probably misguided, the doctor you need may not be any actual doctor — you need one who can perform a memory-retaining brain transplant, for example. The {lo prenu cu tarci} works on any case. the {lo prenu cu sadjo} falls into the vagueness trap by working at the upper end of the range rather than the lower or middle (xorxes does encourage that with his examples, to be sure). Here it would be enough if some few (or many or...) people were Saudis (I am not sure whether one would be enough).

Arnt Richard Johansen wrote:

One of my problems with Jorge's proposal is that it seems to either

conflate two different meanings into one, or have one meaning that is

ill-defined or hard to formalize. What I want is a general algorithm fo=

r

determining whether or not a sentence that uses XS-lo is true or false.

These examples are taken from an IRC discussion I had with xod.

{mi nitcu lo mikce} — obviously true (in the right circumstances)

Here, I take every {mikce} in the world, assign it to {xy.} and for eac=

h

one asks if {mi nitcu xy.} is true. If and only if there is no {xy.} fo=

r

which {mi nitcu xy.} is false, then {mi nitcu lo mikce} is true. No

problem here.

{lo prenu cu tarci} — obviously false

We are then looping through every {prenu}. There are some (actually all=

)

instances in which {xy. tarci} is false, therefore {lo prenu cu tarci} =

is

false. No problem here.

{lo prenu cu sadjo} — obviously true.

We are again looping through every {prenu}. There are some instances in

which {xy. sadjo} is false, therefore {lo prenu cu sadjo} is false. Whi=

ch

is patently wrong, but fits with how I understand XS-lo.

So. What is it that I'm missing here.

--=20

Arnt Richard Johansen http://arj.nvg.org=

/

P=E5 1300-tallet kom tersen. F=F8r og etter det var det meste bare rot,=

men

s=E5 kom Sch=F6nberg og ordnet opp. Puh. Endelig litt system. S=E5 klar=

te Arne

Nordheim =E5 rote det til igjen. — Under Dusken 08/2=

001



Posted by Anonymous on Sat 29 of May, 2004 16:46 GMT

1. Not even that, since his examination is largely irrelevant to the issue: all the doctors he needs may be none of the ones there are.

The claim that all the doctors might be dead and you still need one is about the opaque context generated (covertly in your usage) by {nitcu}: {mi viska lo mikce} won't be true if all the doctors are dead (doctor-corpses not counting as doctors for this case at least).

I suspect that the "just think about it" will not work generally: it does for someof the intensional cases — though not even for all of those.

2: arj's technique is essentially correct here. Look at them all and if none correspond then the {lo} claim is surely false. Since this is probably an analytic claim, thinking about it may be sufficient.

3: Yes, he screws up here: the run through is right but he should be counting hits, not misses. Admittedly, not fixed number of hits is the right one, but cases of no hits will make the whole false (and maybe cases of too few hits — precisely as vague as that seems). But I doubt that thinking about it will help here: Graustarkians are at least as likely a group of people as Saudis coonsidered conceptually but lo prenu are not Graustarkian. You have to check in the real world.

4: And the same to you (see examples). He is merely taking {lo broda} a seriously as you sometimes do, but being consistent about it.


> {mi nitcu lo mikce} — obviously true (in the right circumstances)

> Here, I take every {mikce} in the world, assign it to {xy.} and for eac=

> h

> one asks if {mi nitcu xy.} is true. If and only if there is no {xy.} fo=

> r

> which {mi nitcu xy.} is false, then {mi nitcu lo mikce} is true. No

> problem here.

1. But what you are describing is {mi nitcu ro mikce}.

In {mi nitcu lo mikce} you don't examine any doctor. All you

need to know is what a doctor is. Then, knowing what a doctor

is, you ask yourself, is that what I need? If the answer is yes,

then {mi nitcu lo mikce} is true. You don't have to examine any

doctors. As xod says, all the doctors may be dead and you may

still need one.

> {lo prenu cu tarci} — obviously false

> We are then looping through every {prenu}. There are some (actually all=

> )

> instances in which {xy. tarci} is false, therefore {lo prenu cu tarci} =

> is

> false. No problem here.

2:Again you are describing the procedure to check {ro prenu cu tarci}.

For {lo prenu cu tarci} what you need to know is what a prenu is.

Knowing what a prenu is, do you think it is (at least sometimes)

a tarci? If not, then the sentence is false.

> {lo prenu cu sadjo} — obviously true.

> We are again looping through every {prenu}. There are some instances in

> which {xy. sadjo} is false, therefore {lo prenu cu sadjo} is false. Whi=

> ch

> is patently wrong, but fits with how I understand XS-lo.

3:Again, you are describing {ro prenu cu sadjo}, which as you conclude

is false.

To check {lo prenu cu sadjo} you start from your knowledge of

what a prenu is. Then you ask yourself, is it (at least sometimes)

a sadjo? If yes then {lo prenu cu (su'oroi) sadjo} is true.

{lo prenu roroi fe'eroroi sadjo} is obviously false. People can be

Saudis, but people are not always and everywhere Saudis.

> So. What is it that I'm missing here.

4:You seem to be confusing lo with ro.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 16:46 GMT

pc:

> And

> most of these are peculiar cases (if really cases of {lo} at all):

> generalities, gnomic utterances, maxims and the like – things that are more

> or less universal; that is not {lo} home ground (and very likely not its

> ground at all).

But there is nothing peculiar about these sentences. They are

everyday things people say, and which a fluent Lojban speaker

should be able to produce without a second of hesitation. I found

most of the English sentences with simple Google searches, I did

not make any of them up myself. If such sentences cannot be produced

easily with current Lojban, then there is something wrong with

current Lojban.

Talking about generalities is basic, it happens all the time.

I take it you would use your proposed {xo'o} for many of the examples.

That's a possibility. The disadvantage is that most resulting texts,

which will certainly be full of {xo'o}s because general claims

are very common, will not look like the Lojban that has been produced

in the last twenty years or so. With {lo}, on the other hand, Lojban

will continue to look like so-far-Lojban.

> ei lo verba cu mutce fraxu lo makcu prenu

> Children should always show great forbearance

> toward grown-up people.

It is kind of a maxim, yes. I cannot tell from your words whether

you approve of this translation or whether you would translate it

differently. How would you translate maxims, which are relatively

frequent in any language?

> ku'i uinai mi na viska lo lanme pa'o lo tanxe

> i ju'ocu'i mi milxe simsa lo makcu prenu

> But I, alas, do not see sheep through the walls of

> boxes. Perhaps I am a little like the grown-ups.

>

> I am not sure whether {pa'o} works like this, but the {lo}s in the first

> sentence work out right. A good example (though perhaps for later), since it

> reminds us that universals in negative contexts are expressed existentially:

> “any sheep through any box” (is “the walls of” just a flourish? This eems to

> apply as well to looking through a tubular box lacking both ends.

The original doesn't mention walls: "Mais moi, malheureusement, je ne sais

pas voir les moutons à travers les caisses." I guess context helps make it

clear what is meant: The author has drawn a box, and the little prince

is very happy with the sheep he says is inside of the box. He had rejected

all the previous attempted drawings of sheep for one reason or another.

So at least for negative generic claims you approve of the use of lo.

(I would take {mi na viska su'o lamne pa'o su'o tanxe} to be a more

concrete claim, though.)

> The {lo}

> in the second sentence is probably about a species (etc.) since it is going

> on to some property. I would use {la'e} here, but that is only a reasonable

> start of working out how to talk about species.

But {la'e} still requires another gadri. Do you mean {la'e lo makcu prenu}?

Is that the same as {la'e su'o makcu prenu}?

The only use of {la'e} I know is with {la'e di'u}, to refer to what the

previous sentence says. So {la'e di'u cinri} is "that's interesting", not

the previous sentence itself but what it says. Is that the same {la'e}

that takes you from a grown-up to grown-ups in general?

> ca lo nicte lo cinfo cu kalte lo cidja

> At night lions hunt for food.

I can't tell from your words whether you approve or not of

this translation.

> lo pa pixra cu se vamji lo ki'o valsi

> One picture is worth a thousand words.

>

> Ah, I forgot this aspect of your work with quantifiers. {lo ki'o valsi}

> looks OK and not noticeably different from {ki'o valsi}

But it is noticeably different. The picture is worth the same as the

thousand words together, it is not worth the same as a word 1000 times.

{ki'o valsi} would claim that there are 1000 x which are words, such that

the picture is worth x. So for example, the picture is worth "the", the

picture is worth "little", the picture is worth "house", etc, 1000

times. With {lo ki'o valsi} we are talking of a whole bunch of 1000

words put together.

>– presumably the

> words could be spelled out in each case, maybe several different ways,

> indeed. Presumably this is gnomic again so the first {lo} is either

> universal or about species or perhaps {la’e}.

So what is a Lojban speaker to say?

> de'i li 1960 lo pare sovda cu fepni li 42

> In 1960 a dozen eggs cost 42 cents.

>

> Same old, same old. It was not just one dozen but just about any dozen there

> was – implicit exception in force (crested floo-floo birds’ – now extinct –

> eggs, certified organic, …). Iam inclining more and more to {la’e} here.

Wouldn't that turn {la'e di'u} into generic "sentences like the previous

one", instead of "the referent of the previous sentence"? That's too much

of a change on existing usage, and besides we would need a new way of

doing {la'e di'u}. (In fact, I think it would be great to assign say

{tau} and {tei} to {la'e di'u} and {la'e de'u}, but that's another

thread altogether.)

> lo ctuca cu fendi lo selctu mu lo vo tadni

> The teacher will divide the class

> into five groups of four students.

>

> Hey, some basic cases, though {le ctuca} makes better sense — this seems to

> be a particular occasion. So, come to that, {le selctu} or even {lei selctu}.

> But the {mu lo vo tadni} is nice.

The English sentence can be generic too, and in context it was:

"LESSON SUMMARY: 30 minutes

The teacher will ask the students to brainstorm ideas for each column.

For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim…)

The teacher will record an answer on each piece of oak tag in the column.

The teacher will divide the class into five groups of four students.

The teacher will take all of the oak tag pieces and place them face down

in groups according to characters, setting, and problem. The teacher will

ask each group to choose one piece of oak tag from each group. ..."

It is not about some particular teacher or class that the speaker

has in mind. It is more general.

> lo bidjylinsi pe lo ze seldri cu se pagbu

> ze lo ze bidju e ji'a ci lo pa bidju e lo kucysni

> The Rosary of the Seven Sorrows consists of seven groups

> of seven beads, with three additional beads and a Crucifix.

>

> This looks like it is about a certain class of things, a particular kind of

> rosary (and, indeed, if it was about a unique thing {le} would be

> appropriate. Here there is none of the worry about exceptions that the more

> gnomic cases call for, so this could be done with {ro}. But I take it to be

> about the kind, laying out its particularities. In that case, the last three

> {lo}s are just any-olds; put them together in this way and you get a rosaary

> of the right sort. The first should be for species or kind and whether this

> form or some other covers these cases I leave for a while.

I'll take that as semi-approved then.

> lo sanli darxi bo dakli cu culno lo djacu onai lo canre to lo djacu

> cu pukmau ki'u lo nu slilu tolcando toi gi'e bunda li ji'i 270

> Standing punching bags are filled with water or sand - water

> being preferable because of the wave-motion created - and

> weigh about 270lbs.

>

> Species substance substance substance species (but maybe, in all this scope,

> {lo} would work)

Do you approve or disapprove of using {lo} for substance?

> lo pavyseljirna cu ranmi danlu gi'e catlu lo ka ge ce'u xirma

> gi lo jirna cu cpana lo sedycra be ce'u

> Unicorns are mythical creatures that look like a horse

> with a horn coming out of their foreheads.

>

> Species conventional (could be {le} just as well) ok conventional (but I

> think {le} is a more sensible convention). This looks like a good safish way

> to talk about species (well, with the appropriate gadri, of course).

I can't tell whether you approve or not.

> bilga lenu jdice lenu roroi pilno lo mokla tirxe

> (to zoigy. velar gy. toi) jonai crane (to zoigy.

> alveolar gy. toi)

> tavla fi le tutra pe le terdi

>

> I’m not sure about the context here, but this looks ok: on each occasion one

> use some velar (or alveolar). But complex for the point. How is this a

> problem solved; it seems to be basic {lo}

He meant to say that we should pick either velar or alveolar for all

occasions.

> What is the role of the blue

> expressions?

They are links to the page where the sentence was taken from, so you

can check the context if you want.

> le cmana lo cidja ba claxu

> In the mountains there is no food.

> lapoi pelxu ku'o trajynobli

>

> Normal usage – well it is good to see that implicit negation works like

> ex-lciti (but does it? I hope so).

Why not analyse {nitcu} as an implicit negation too, then?

> le dargu pe lo xamgu bangu cu kargu

> The road of the good language is costly.

> lapoi pelxu ku'o trajynobli

>

> Specific or universal (probably the latter — it seems merely factual)

You agree with me that it is not equivalent to {le dargu pe su'o xamgu

bangu} then.

> la jyryr. tolkien. cu te cukta la djine turni (to la'o

> gy Lord of the Rings gy toi) .e le so'omoi be lo

> xanri munje lisri ca le lampru na'acto

> tenguar

>

> Species or set (probably the latter). “the severalth” is nice, though not a

> clear as it might be; I suppose it is to me “one of several” or just “pretty

> far along in the set ordered by … (date?)”

Maybe he meant {so'omei}.

Either way, {lo xanri munje lisri} seems to me generic. {su'o} would

not make sense there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Re: BPFK Section: gadri


Posted by xorxes on Sat 29 of May, 2004 16:57 GMT posts: 1912

(Please everybody, use the 'discuss' button and not the

'comments' button to make comments and proposals on the

contents of the page. It makes it easier to reply, thank you.)

pier:

> {lo pare sovda cu fepni li 42} to me means that there are 12

> eggs, each of which costs 42 cents. If a dozen eggs costs 42

> cents, I'd say {lo sovda paremei cu fepni li 42} or maybe {loi

> pare sovda cu fepni li 42}.

{pare sovda cu fepni li 42} says that there are 12 eggs, each

of which costs 42 cents. {lo pare sovda} refers to a twelvesome

of eggs. It is like {lo sovda paremei} but without the tanru

imprecision. {loi pare sovda} is indeed equivalent to

{lo pare sovda}.

> {lo'e se} with the word for a kind of organism is a special case

> and should be mentioned. It refers to the taxon containing all

> members of the kind and no others, if that exists. So {lo'e se

> guzme} means the family Cucurbitaceae, while {lo se guzme}

>can be Cucumis, Sicyos, Luffa, or any of several others. {lo'e se

> jesymabru}, however, has no clear meaning, as {jesymabru}

> can refer to both spiny anteaters and hedgehogs.

I have added a comment to that effect. If others have any

objections to this, please speak up now or forever hold your

peace.

mu'o mi'e xorxes



Posted by Anonymous on Sat 29 of May, 2004 19:14 GMT

1. Quite true and Lojban needs to have ways to say them,. The question is whether thye are proper for the historical continuity of {lo} and whether one concept can cover them all. I think that the answer is at least doubtful on the first and even more so one the second. They are peculiar only against standard {lo} and thus as generalization away from that. As for there being something wrong with current Lojban, we have known that for as many years as there has been current Lojban and have tried to fix it countlless times. There have been many good ideas about how to do it, but no set of them (no one suggestion has solved all — or even most — of the problems) has been accepted and what seems to gain ocasional favor is a good partial solution pushed to absurdity. Until it collapses and we go back to nearly square one with all the intervening suggestions (and even what was good in the popular suggestion) lost.

2. Well, not basic basic but very common indeed and the more so as we get more remote from nature and live in conceptual realms. If this is what Lojban so far has been doing with {lo} (as I take is your point) then to that extent it is bad Lojban and ought not to be preserved as exemplary. It sounds as though the fact that people were not called out on those uses of {lo} in the past kept them from looking for legitimate ways to speak in generalities. Admittedly, much of what the generalities brought up here say is pretty compatible with {lo}, covering the gap between none and all, although in a rather different way. As noted, one case is usually too few, all is closer but with a variety of ad hoc exceptions, and — though I have not mentioned it before — often with some cases counting more than others (absolute monarch are better for figuring out what kings do than constitutional monarchs, for example). So in factthese notions do not belong in the same area with those where

counting is to the point. But {lo} just does belong in that area and so ought not be brought into this discussion. It may turn out that, on zipfy gounds for example, we want to start to use {lo} in this way and a longer expression for old {lo} (arguably, {su'o} already works), but that is a major decision, not to be made casually and without comment to speak of on a wiki page (and so looking remarkably like a done deal). I don't, by the way, think you consistently use {lo} in this way, but that is another matter at this point.

3: Depending on context, I mght do either of two things or maybe something else. If this is just being laid down sententiously, I would probably take all the {lo}s as {ro}s and leave it to casuistry to deal with the exceptions (there always are some; only a person with no life — like Kant — believes in absolute moral generalizations and no one that I can find has ever seriously tried to list all the exceptions beforehand for any rule). On the other hand, if this is admonitory — codger to irksome kids, I might go for brass tacks and deal with "you whippersnappers" and "me, your older, wiser and better" {do noi malverba} and {mi noi makcu}. If I was being cagey I might flavor the universal form with an "as a rule" modal/tense/something (which we lack but need for other things).

4: Negations are a nice source for confusions: if no x is y, than it is not the case that at least one is nor is it the case that some unspecified/ble number are. Ditto if we really mean "all" (which we rarely do, of course).

5: Sorry; misprint. {lo'e}

6: I suspect that I meant that this was basically untranslatable into current Lojban. It looks like a general claim but can't be universal, since there are exceptions and no ground for casuistry. On the other hand, clearly more than one lion and one night is meant. I think that the reference to food should probably be to substance not to particulars or generals so, even if {lo} turns out to be OK for the first two, it is not for the last.

7: Thanx. I doing get this usage completely yet, though I think it is a good one. It is a shift, however, and so needs more discussion (and a large warning) on the wiki page.

8: I am inclined to use {lo'e pixra} and {lo or su'o ki'o valsi}. This is relatively particular, one picture, one set of 1000 words, so full generic does not seem to apply as would "Pictures are worth thousands of words each," say. Here the inspecificity of the generic usage is further complicated. On a bet, not only are some {unspecified number of) pictures worth a thousand words, but 1000 is merely a round number of unspecified import: some are worth only 950, some 1200, quite a few less than 100, a rare few upwards toward millions. A nice "roughly speaking" modal might help here. (I know it is improper to take this stuff literally — what, as the poet asked, are words worth after all. But the serious cases of this sort are also very common — something about eggs coming up.)

9: I see that the {la'e} for {lo'e} runs through this whole thing. Sorry again (especially since I never seem to have backed it up with a "the typical"). In this case, since there are statistics (on a bet) The average — in some sense (though they probably all three coincide).

10: Ahah! Context makes a difference; it looked like a report but it was a direction ({e'u} or {e'o} or {ei} or maybe something more complex). Still, as read by each particular teacher and applied in a particular classroom, it is quite particular, so {le ctuca} and {le [[or%20%7Blei%7D|or {lei}]] selctu}.

11: Nicely put. This is sort of a definition (lacking something about stringing them together in a loop with the last set dangling from the join and about how the groups are demarcated), I am not sure how to do this — even if we have what I have been calling a generic gadri — but I am sure that it does not require more new stuff. Of course, it could be literally a definition, defining the expression, but that — thouggh it has a long history as an out — does not seem quite fair.

12. Even if {lo} is generic in the sense set out here. To be sure, for the uncountables (in English), {lo djacu} comes pretty close to being about the substance in extension at least. But that doesn't work for countables lo bakni are cattle, not beef.

13. Me either. We've been around about how to say that non-existents don't exist, so I'll leave that part out. I never feel comfortable with whichever gadri I use with {ka} and the like, but one seems a good as the other. {lo jirna} is surely correct, even with old {lo} and I would say {le sedycra} since it is the particular one of the particular unicorn we have got to in applying this property.

14. Yes, though {bilga} that way looks odd. The point is that {lo} is in the scope of {roroi} In that sense I am not sure that this says what it is supposed to. It seems to limit the choices one has to velars and alveolars but not to require the same one all the time. One has decided, apparently, never to use dentals or labials or gutterals or palatals. I haven't a clue at the moment how to clean it up. But the {lo}s are OK.

15: Where is the implicit negation in {nitcu}? To be sure, needing implies lacking; but it does not assert it.

16: Yes indeed. I now would incline to generic (though I wonder about the {le} in that context.)

17: {so'omei} makes not sense in the context. {lo'i} would be nice and safe or maybe even {ro}, but {su'o} clearly does not work here. The generic doesn't very well either, since I think it means we are to run through all of them (with conditions — e.g., unpublished ones, privately printed runs of thirty and the like).

pc:

> And

> most of these are peculiar cases (if really cases of {lo} at all):

> generalities, gnomic utterances, maxims and the like – things that are more

> or less universal; that is not {lo} home ground (and very likely not its

> ground at all).

1.But there is nothing peculiar about these sentences. They are

everyday things people say, and which a fluent Lojban speaker

should be able to produce without a second of hesitation. I found

most of the English sentences with simple Google searches, I did

not make any of them up myself. If such sentences cannot be produced

easily with current Lojban, then there is something wrong with

current Lojban.

2.Talking about generalities is basic, it happens all the time.

I take it you would use your proposed {xo'o} for many of the examples.

That's a possibility. The disadvantage is that most resulting texts,

which will certainly be full of {xo'o}s because general claims

are very common, will not look like the Lojban that has been produced

in the last twenty years or so. With {lo}, on the other hand, Lojban

will continue to look like so-far-Lojban.

> ei lo verba cu mutce fraxu lo makcu prenu

> Children should always show great forbearance

> toward grown-up people.

3.It is kind of a maxim, yes. I cannot tell from your words whether

you approve of this translation or whether you would translate it

differently. How would you translate maxims, which are relatively

frequent in any language?

> ku'i uinai mi na viska lo lanme pa'o lo tanxe

> i ju'ocu'i mi milxe simsa lo makcu prenu

> But I, alas, do not see sheep through the walls of

> boxes. Perhaps I am a little like the grown-ups.

>

> I am not sure whether {pa'o} works like this, but the {lo}s in the first

> sentence work out right. A good example (though perhaps for later), since it

> reminds us that universals in negative contexts are expressed existentially:

> “any sheep through any box” (is “the walls of” just a flourish? This eems to

> apply as well to looking through a tubular box lacking both ends.

The original doesn't mention walls: "Mais moi, malheureusement, je ne sais

pas voir les moutons à travers les caisses." I guess context helps make it

clear what is meant: The author has drawn a box, and the little prince

is very happy with the sheep he says is inside of the box. He had rejected

all the previous attempted drawings of sheep for one reason or another.

4:So at least for negative generic claims you approve of the use of lo.

(I would take {mi na viska su'o lamne pa'o su'o tanxe} to be a more

concrete claim, though.)

> The {lo}

> in the second sentence is probably about a species (etc.) since it is going

> on to some property. I would use {la'e} here, but that is only a reasonable

> start of working out how to talk about species.

5:But {la'e} still requires another gadri. Do you mean {la'e lo makcu prenu}?

Is that the same as {la'e su'o makcu prenu}?

The only use of {la'e} I know is with {la'e di'u}, to refer to what the

previous sentence says. So {la'e di'u cinri} is "that's interesting", not

the previous sentence itself but what it says. Is that the same {la'e}

that takes you from a grown-up to grown-ups in general?

> ca lo nicte lo cinfo cu kalte lo cidja

> At night lions hunt for food.

6:I can't tell from your words whether you approve or not of

this translation.

> lo pa pixra cu se vamji lo ki'o valsi

> One picture is worth a thousand words.

>

> Ah, I forgot this aspect of your work with quantifiers. {lo ki'o valsi}

> looks OK and not noticeably different from {ki'o valsi}

7:But it is noticeably different. The picture is worth the same as the

thousand words together, it is not worth the same as a word 1000 times.

{ki'o valsi} would claim that there are 1000 x which are words, such that

the picture is worth x. So for example, the picture is worth "the", the

picture is worth "little", the picture is worth "house", etc, 1000

times. With {lo ki'o valsi} we are talking of a whole bunch of 1000

words put together.

>– presumably the

> words could be spelled out in each case, maybe several different ways,

> indeed. Presumably this is gnomic again so the first {lo} is either

> universal or about species or perhaps {la’e}.

8: So what is a Lojban speaker to say?

> de'i li 1960 lo pare sovda cu fepni li 42

> In 1960 a dozen eggs cost 42 cents.

>

> Same old, same old. It was not just one dozen but just about any dozen there

> was – implicit exception in force (crested floo-floo birds’ – now extinct –

> eggs, certified organic, …). Iam inclining more and more to {la’e} here.

9:Wouldn't that turn {la'e di'u} into generic "sentences like the previous

one", instead of "the referent of the previous sentence"? That's too much

of a change on existing usage, and besides we would need a new way of

doing {la'e di'u}. (In fact, I think it would be great to assign say

{tau} and {tei} to {la'e di'u} and {la'e de'u}, but that's another

thread altogether.)

> lo ctuca cu fendi lo selctu mu lo vo tadni

> The teacher will divide the class

> into five groups of four students.

>

> Hey, some basic cases, though {le ctuca} makes better sense — this seems to

> be a particular occasion. So, come to that, {le selctu} or even {lei selctu}.

> But the {mu lo vo tadni} is nice.

The English sentence can be generic too, and in context it was:

10:"LESSON SUMMARY: 30 minutes

The teacher will ask the students to brainstorm ideas for each column.

For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim…)

The teacher will record an answer on each piece of oak tag in the column.

The teacher will divide the class into five groups of four students.

The teacher will take all of the oak tag pieces and place them face down

in groups according to characters, setting, and problem. The teacher will

ask each group to choose one piece of oak tag from each group. ..."

It is not about some particular teacher or class that the speaker

has in mind. It is more general.

> lo bidjylinsi pe lo ze seldri cu se pagbu

> ze lo ze bidju e ji'a ci lo pa bidju e lo kucysni

> The Rosary of the Seven Sorrows consists of seven groups

> of seven beads, with three additional beads and a Crucifix.

>

> This looks like it is about a certain class of things, a particular kind of

> rosary (and, indeed, if it was about a unique thing {le} would be

> appropriate. Here there is none of the worry about exceptions that the more

> gnomic cases call for, so this could be done with {ro}. But I take it to be

> about the kind, laying out its particularities. In that case, the last three

> {lo}s are just any-olds; put them together in this way and you get a rosaary

> of the right sort. The first should be for species or kind and whether this

> form or some other covers these cases I leave for a while.

11:I'll take that as semi-approved then.

> lo sanli darxi bo dakli cu culno lo djacu onai lo canre to lo djacu

> cu pukmau ki'u lo nu slilu tolcando toi gi'e bunda li ji'i 270

> Standing punching bags are filled with water or sand - water

> being preferable because of the wave-motion created - and

> weigh about 270lbs.

>

> Species substance substance substance species (but maybe, in all this scope,

> {lo} would work)

12:Do you approve or disapprove of using {lo} for substance?

> lo pavyseljirna cu ranmi danlu gi'e catlu lo ka ge ce'u xirma

> gi lo jirna cu cpana lo sedycra be ce'u

> Unicorns are mythical creatures that look like a horse

> with a horn coming out of their foreheads.

>

> Species conventional (could be {le} just as well) ok conventional (but I

> think {le} is a more sensible convention). This looks like a good safish way

> to talk about species (well, with the appropriate gadri, of course).

13.I can't tell whether you approve or not.

> bilga lenu jdice lenu roroi pilno lo mokla tirxe

> (to zoigy. velar gy. toi) jonai crane (to zoigy.

> alveolar gy. toi)

> tavla fi le tutra pe le terdi

>

> I’m not sure about the context here, but this looks ok: on each occasion one

> use some velar (or alveolar). But complex for the point. How is this a

> problem solved; it seems to be basic {lo}

14.He meant to say that we should pick either velar or alveolar for all

occasions.

> What is the role of the blue

> expressions?

They are links to the page where the sentence was taken from, so you

can check the context if you want.

> le cmana lo cidja ba claxu

> In the mountains there is no food.

> lapoi pelxu ku'o trajynobli

>

> Normal usage – well it is good to see that implicit negation works like

> ex-lciti (but does it? I hope so).

15: Why not analyse {nitcu} as an implicit negation too, then?

> le dargu pe lo xamgu bangu cu kargu

> The road of the good language is costly.

> lapoi pelxu ku'o trajynobli

>

> Specific or universal (probably the latter — it seems merely factual)

16: You agree with me that it is not equivalent to {le dargu pe su'o xamgu

bangu} then.

> la jyryr. tolkien. cu te cukta la djine turni (to la'o

> gy Lord of the Rings gy toi) .e le so'omoi be lo

> xanri munje lisri ca le lampru na'acto

> tenguar

>

> Species or set (probably the latter). “the severalth” is nice, though not a

> clear as it might be; I suppose it is to me “one of several” or just “pretty

> far along in the set ordered by … (date?)”

17. Maybe he meant {so'omei}.

Either way, {lo xanri munje lisri} seems to me generic. {su'o} would

not make sense there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 20:01 GMT

1. I suppose {paremei} is strictly a tanru, but it is hard to see how it could be more precise, I like {lo pare sovda} better for all that.

2: I gather that {lo'e} is in play as well as {lo}, moving from "the typical" to "the taxon" or so (probably not literal Linnean taxa only but that sort of thing on any informal level). I think that talk about that sort of thing usually is just generic "cucumbers do thus and so", meaning more than some, probably not all and certainly the ones that I am fond of, pretty much what xorxes has been using {lo} for most of the time. But this case nicely mmuddles things, since the critters about which we are talking are exactly subtaxa, not their representatives. No problems with that, actually, but some with the first part, the taxon itself. We are set up for talking about members (etc.) not the abstracts. But I said we needed a device for these and here finally is a case — I think.

[email protected] wrote:

Re: BPFK Section: gadri

(Please everybody, use the 'discuss' button and not the

'comments' button to make comments and proposals on the

contents of the page. It makes it easier to reply, thank you.)

pier:

> {lo pare sovda cu fepni li 42} to me means that there are 12

> eggs, each of which costs 42 cents. If a dozen eggs costs 42

> cents, I'd say {lo sovda paremei cu fepni li 42} or maybe {loi

> pare sovda cu fepni li 42}.

{pare sovda cu fepni li 42} says that there are 12 eggs, each

of which costs 42 cents. {lo pare sovda} refers to a twelvesome

of eggs.!. It is like {lo sovda paremei} but without the tanru

imprecision. {loi pare sovda} is indeed equivalent to

{lo pare sovda}.

2:> {lo'e se} with the word for a kind of organism is a special case

> and should be mentioned. It refers to the taxon containing all

> members of the kind and no others, if that exists. So {lo'e se

> guzme} means the family Cucurbitaceae, while {lo se guzme}

>can be Cucumis, Sicyos, Luffa, or any of several others. {lo'e se

> jesymabru}, however, has no clear meaning, as {jesymabru}

> can refer to both spiny anteaters and hedgehogs.

I have added a comment to that effect. If others have any

objections to this, please speak up now or forever hold your

peace.

mu'o mi'e xorxes



Posted by Anonymous on Sat 29 of May, 2004 20:01 GMT

pc:

> 1. I suppose {paremei} is strictly a tanru, but it is hard to see how it

> could be more precise, I like {lo pare sovda} better for all that.

{paremei} is not the tanru, {sovda paremei} is. An egg type of dozen

is most likely but not necessarily a dozen eggs.

> 2: I gather that {lo'e} is in play as well as {lo}, moving from "the typical"

> to "the taxon" or so (probably not literal Linnean taxa only but that sort of

> thing on any informal level). I think that talk about that sort of thing

> usually is just generic "cucumbers do thus and so", meaning more than some,

> probably not all and certainly the ones that I am fond of, pretty much what

> xorxes has been using {lo} for most of the time. But this case nicely

> mmuddles things, since the critters about which we are talking are exactly

> subtaxa, not their representatives. No problems with that, actually, but some

> with the first part, the taxon itself. We are set up for talking about

> members (etc.) not the abstracts. But I said we needed a device for these

> and here finally is a case — I think.

I'm not sure if you're taking into account the "se" here.

The x1 of guzme is for the cucumbers and the x2 for the taxon

or whatever. Pierre is talking about {lo'e se guzme}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 23:30 GMT

pc:

> As for there being

> something wrong with current Lojban, we have known that for as many years as

> there has been current Lojban and have tried to fix it countlless times.

Not sure who you mean by "we", but some people think that nothing

should be touched.

> It

> sounds as though the fact that people were not called out on those uses of

> {lo} in the past kept them from looking for legitimate ways to speak in

> generalities.

People were called out on those uses all the time, you can see

that from the list archives. The problem is that we could say

{su'o lo} was wrong but we could not say what was right instead.

Any way of generic speaking could be objected to. And {lo} is not

the only thing people use for lack of something better. {le} and

{loi} are also popular alternatives.

> It may turn out that, on zipfy

> gounds for example, we want to start to use {lo} in this way and a longer

> expression for old {lo} (arguably, {su'o} already works), but that is a major

> decision, not to be made casually and without comment to speak of on a wiki

> page (and so looking remarkably like a done deal).

The wiki page is clearly labeled as a proposal, and it is clearly

part of the BPFK work we are doing. This is a proposal to be

discussed, amended as necessary, and voted on.

> 10: Ahah! Context makes a difference; it looked like a report but it was a

> direction ({e'u} or {e'o} or {ei} or maybe something more complex). Still,

> as read by each particular teacher and applied in a particular classroom, it

> is quite particular, so {le ctuca} and {le [[or%20%7Blei%7D|or {lei}]] selctu}.

So even though the speaker does not have any particular

teacher in mind, you think he should use {le} because some teacher

reading it might have a particular one in mind? What about other

readers that may simply be interested in teaching methods but not

in actually performing this particular lesson?

> 12. Even if {lo} is generic in the sense set out here. To be sure, for the

> uncountables (in English), {lo djacu} comes pretty close to being about the

> substance in extension at least. But that doesn't work for countables lo

> bakni are cattle, not beef.

{lo tu'o gerku} is proposed for the dog all over the pavement.

"Beef" however is probably better as {bakni rectu}.

({ractu rectu} for rabbit.)

> 15: Where is the implicit negation in {nitcu}? To be sure, needing implies

> lacking; but it does not assert it.

Does {claxu} assert not having, or does it just imply it?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 23:30 GMT

1. Oops! Thinking "lujvo" when reading "tanru."

2. Yes, that is what makes it so interesting. Apparently he wants the things which generally do so and so to be the subtaxa — maybe species, maybe genera but certainly not the indivdual plants — that fall under the whatever it is. So there is some need for a marker for these things.

Jorge Llambías wrote:

pc:

> 1. I suppose {paremei} is strictly a tanru, but it is hard to see how it

> could be more precise, I like {lo pare sovda} better for all that.

1. {paremei} is not the tanru, {sovda paremei} is. An egg type of dozen

is most likely but not necessarily a dozen eggs.

> 2: I gather that {lo'e} is in play as well as {lo}, moving from "the typical"

> to "the taxon" or so (probably not literal Linnean taxa only but that sort of

> thing on any informal level). I think that talk about that sort of thing

> usually is just generic "cucumbers do thus and so", meaning more than some,

> probably not all and certainly the ones that I am fond of, pretty much what

> xorxes has been using {lo} for most of the time. But this case nicely

> mmuddles things, since the critters about which we are talking are exactly

> subtaxa, not their representatives. No problems with that, actually, but some

> with the first part, the taxon itself. We are set up for talking about

> members (etc.) not the abstracts. But I said we needed a device for these

> and here finally is a case — I think.

2>I'm not sure if you're taking into account the "se" here.

The x1 of guzme is for the cucumbers and the x2 for the taxon

or whatever. Pierre is talking about {lo'e se guzme}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT

Judging from what Nick wrote at

http://www.lojban.org/tiki/3Dgadri+report%2C+aug+2003

this is a field of active research in linguistics/philosophy, and we

should not expect to arrive at a consensus solution here. To do so would

mean that we either had broken significant ground worthy of publishing,

or more likely, that we had succeeded in deluding ourselves. The BF

commissioners should attempt to clarify the situation beyond its current

state, but not expect a solution that lies beyond all criticism. ju'a

..e'unai lo prane cu bradi lo xamgu

Jorge Llamb=EDas wrote:

>pc:

> =20

>

>>And

>>most of these are peculiar cases (if really cases of {lo} at all):

>>generalities, gnomic utterances, maxims and the like =96 things that ar=

e more

>>or less universal; that is not {lo} home ground (and very likely not it=

s

>>ground at all).

>> =20

>>

>

>But there is nothing peculiar about these sentences. They are

>everyday things people say, and which a fluent Lojban speaker

>should be able to produce without a second of hesitation. I found

>most of the English sentences with simple Google searches, I did=20

>not make any of them up myself. If such sentences cannot be produced=20

>easily with current Lojban, then there is something wrong with=20

>current Lojban.

>

>Talking about generalities is basic, it happens all the time.

>I take it you would use your proposed {xo'o} for many of the examples.

>That's a possibility. The disadvantage is that most resulting texts,

>which will certainly be full of {xo'o}s because general claims

>are very common, will not look like the Lojban that has been produced

>in the last twenty years or so. With {lo}, on the other hand, Lojban

>will continue to look like so-far-Lojban.=20

>

> =20

>

>>ei lo verba cu mutce fraxu lo makcu prenu

>>Children should always show great forbearance

>>toward grown-up people.

>> =20

>>

>It is kind of a maxim, yes. I cannot tell from your words whether=20

>you approve of this translation or whether you would translate it=20

>differently. How would you translate maxims, which are relatively=20

>frequent in any language?=20

> =20

>

It seems to me that maxims, and claims that sweep similarly broad,

should be translated with ro, not lo. And since ro applies to every

member, and hence to the whole type by definition, then in cooperative

usage lo will be used to discuss subsets, often but not necessarily prope=

r.

>>ku'i uinai mi na viska lo lanme pa'o lo tanxe

>>i ju'ocu'i mi milxe simsa lo makcu prenu

>>But I, alas, do not see sheep through the walls of

>>boxes. Perhaps I am a little like the grown-ups.

>>=20

>>I am not sure whether {pa'o} works like this, but the {lo}s in the firs=

t

>>sentence work out right. A good example (though perhaps for later), si=

nce it

>>reminds us that universals in negative contexts are expressed existenti=

ally:

>>=93any sheep through any box=94 (is =93the walls of=94 just a flourish?=

This eems to

>>apply as well to looking through a tubular box lacking both ends.=20

>> =20

>>

>

>The original doesn't mention walls: "Mais moi, malheureusement, je ne sa=

is

>pas voir les moutons =E0 travers les caisses." I guess context helps mak=

e it=20

>clear what is meant: The author has drawn a box, and the little prince

>is very happy with the sheep he says is inside of the box. He had reject=

ed

>all the previous attempted drawings of sheep for one reason or another.

>So at least for negative generic claims you approve of the use of lo.

>(I would take {mi na viska su'o lamne pa'o su'o tanxe} to be a more

>concrete claim, though.)

>=20

> =20

>

>>The {lo}

>>in the second sentence is probably about a species (etc.) since it is g=

oing

>>on to some property. I would use {la'e} here, but that is only a reaso=

nable

>>start of working out how to talk about species.

>> =20

>>

>

>But {la'e} still requires another gadri. Do you mean {la'e lo makcu pren=

u}?

>Is that the same as {la'e su'o makcu prenu}?=20

>

>The only use of {la'e} I know is with {la'e di'u}, to refer to what the=20

>previous sentence says. So {la'e di'u cinri} is "that's interesting", no=

t=20

>the previous sentence itself but what it says. Is that the same {la'e}=20

>that takes you from a grown-up to grown-ups in general?

>

> =20

>

>>ca lo nicte lo cinfo cu kalte lo cidja

>>At night lions hunt for food.

>> =20

>>

>

>I can't tell from your words whether you approve or not of

>this translation.

> =20

>

This is a quasi-definitional sentence such as we might expect to find in

an encyclopedia. Hence, I suggest {ca lo nicte ro cinfo cu kalte lo

cidja}, with possible shuffling if needed to avoid scope side effects.

Such a claim is a universal claim, not simply a non-specific one. {ca lo

nicte lo cinfo cu kalte lo cidja} should not be interpreted as general

claim about lions any more than it's a general claim about nights or

food. (It should be clear that the treatment of lion in that sentence

should be tagged differently than nights and food.) If you want to

wiggle out of making an absolute claim refuted by a single wacky lion,

so'a cinfo will do.

>>lo pa pixra cu se vamji lo ki'o valsi

>>One picture is worth a thousand words.

>>

>>Ah, I forgot this aspect of your work with quantifiers. {lo ki'o valsi=

}

>>looks OK and not noticeably different from {ki'o valsi}=20

>> =20

>>

>

>But it is noticeably different. The picture is worth the same as the

>thousand words together, it is not worth the same as a word 1000 times.=20

>{ki'o valsi} would claim that there are 1000 x which are words, such tha=

t=20

>the picture is worth x. So for example, the picture is worth "the", the=20

>picture is worth "little", the picture is worth "house", etc, 1000=20

>times. With {lo ki'o valsi} we are talking of a whole bunch of 1000=20

>words put together.=20

> =20

>

Here is another case for ro pixra. Use pe'a as nerd-proofing, lest some

lamer produce a picture worth only 999 words.

>>=96 presumably the

>>words could be spelled out in each case, maybe several different ways,

>>indeed. Presumably this is gnomic again so the first {lo} is either

>>universal or about species or perhaps {la=92e}.

>> =20

>>

>

>So what is a Lojban speaker to say?

>

> =20

>

>>de'i li 1960 lo pare sovda cu fepni li 42

>>In 1960 a dozen eggs cost 42 cents.

>>

>>Same old, same old. It was not just one dozen but just about any dozen=

there

>>was =96 implicit exception in force (crested floo-floo birds=92 =96 now=

extinct =96

>>eggs, certified organic, =85). Iam inclining more and more to {la=92e}=

here.

>> =20

>>

>

>Wouldn't that turn {la'e di'u} into generic "sentences like the previous=

=20

>one", instead of "the referent of the previous sentence"? That's too muc=

h

>of a change on existing usage, and besides we would need a new way of

>doing {la'e di'u}. (In fact, I think it would be great to assign say

>{tau} and {tei} to {la'e di'u} and {la'e de'u}, but that's another=20

>thread altogether.)

>

> =20

>

>>lo ctuca cu fendi lo selctu mu lo vo tadni

>>The teacher will divide the class

>>into five groups of four students.

>>

>>Hey, some basic cases, though {le ctuca} makes better sense — this se=

ems to

>>be a particular occasion. So, come to that, {le selctu} or even {lei se=

lctu}.

>> But the {mu lo vo tadni} is nice.

>> =20

>>

>

>The English sentence can be generic too, and in context it was:

>

>"LESSON SUMMARY: 30 minutes

>

>The teacher will ask the students to brainstorm ideas for each column.=20

>For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim=85)

>

>The teacher will record an answer on each piece of oak tag in the column=

..

>

>The teacher will divide the class into five groups of four students.

>

>The teacher will take all of the oak tag pieces and place them face down=

=20

>in groups according to characters, setting, and problem. The teacher wil=

l=20

>ask each group to choose one piece of oak tag from each group. ..."

>

>It is not about some particular teacher or class that the speaker=20

>has in mind. It is more general.

> =20

>

It could be argued that the author is writing a script and has a

particular scene in mind, and in that sense is referring to that

specific teacher. I would expect the all-but-first references to use le;

the first reference having grabbed a random teacher out of the air, and

the following references referring to that teacher and only that one.

--=20

Motorists honked in celebration in this Ramadi as news spread of the=20

assassination of the president of the Iraqi Governing Council Ezzidin Sal=

im=20

Monday. "The GC is nothing," one man shouted. "They are not the Governing=

=20

Council. They are the Prostitution Council."



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT

Arnt Richard Johansen wrote:

>One of my problems with Jorge's proposal is that it seems to either

>conflate two different meanings into one, or have one meaning that is

>ill-defined or hard to formalize. What I want is a general algorithm fo=

>r

>determining whether or not a sentence that uses XS-lo is true or false.

>

>

If you check out

http://www.lojban.org/tiki/gadri+report%2C+aug+2003

and search for the section "Problem 4. Intensionals." you'll see that by

their nature intensionals cannot be enumerate, and so evade the precise

analysis you're looking for. That is not a solvable flaw, but a feature

of what intensionals are.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT

1. After going to meetings and reading list for 30 years, I can assure that — whatever they say publicaly — every person who has spent enough time on Loglan or Lojban to feel up to talking about it has something they want to change. Some many things or broad changes, some details, almost everyone additions.

2: There have been no shortage of alternatives proposed, just none that got the right sort of support behind it (I think I have made three and & at least two others — but those cancel eachother out, of course joke). The problems with {le} (which is a non-starter) and {loi} (which does one small part of the work in some contexts) is that they, like {lo}, already have clear uses prescribed. I think that too many people have thought that Lojban was set in concrete or that no proposal would be accepted (because there were those in power who held that nothing may change — a holdover from Loglan, where it was often true, unless you could convice JCB that he thought of it, which did happen occasionally). So the slogged on with makeshifts rather than getting together a good case: examples, clear explanation why nothing currently works, clear explanation of how the proposed extension works, estimates of cost and advantage and so on. That is a lot of work for one person to do and

loCCan has not been very good at creating committees that actually do things (what gets done gets done by one person doing it). My comment was an attempt to shortcut the process slightly: after fifty years of carping it is clear that there are something which we want to say but which all are attempts to say in current language have ended in failure. Let's just create ways to say them and get on with it. If we do figure out how to say them without all the additions (and once they get said a lot that is a real possibility) then we can drop the additons and retrofit the text corpus. Since at least 1976, when I started seeing the carping, claims fuzzily between all and some, claims about the substance of things, and claims about nodes in the great conceptual tree (not usually put that way, to be sure) have come along annually (if not daily). I take that as enough data, let's fix it.

3. I know how it is labelled and I know the effect of seeing something in official looking print, especially with the kind of power that BPFK appears to wield. That it is not a done deal or even close can't be emphasized too often (actually, I suppose it could: if it gave the impression that nothing was ever done and so there was either no point in trying or that one might as well toss everything in one's head into the hopper since it is all persiflage anyhow).

4: I don't think that a teacher reading it for guidance has anydoubt who the writer has in mind: the teacher reading this copy for guidance. The casual reader doesn't either: the teacher who is using the guidelines in an actual situation. This is a perhaps metaphorical sense of in mind but the point is that, in a given case, the teacher meant is always specific, not just any old teacher. The whole could be framed differently, as a report of what went on in a (according to the authors) well-run classroom or as a general direction for how a classroom ought to be run, but this is direction for how you the student teacher are to go.

5: But number may not be irrelevant here, one is quite capable of being interested in reporting a smear of two-dogs on the highway. I forget all the proposals and all the cases for and against; I just note this has been a problem over the years, so let's fix it, The {rectu} only works for the (mainly) edible parts of critters, not for goo.

6: Asserts.

Jorge Llambías wrote:

pc:

> As for there being

> something wrong with current Lojban, we have known that for as many years as

> there has been current Lojban and have tried to fix it countlless times.

1.Not sure who you mean by "we", but some people think that nothing

should be touched.

> It

> sounds as though the fact that people were not called out on those uses of

> {lo} in the past kept them from looking for legitimate ways to speak in

> generalities.

2:People were called out on those uses all the time, you can see

that from the list archives. The problem is that we could say

{su'o lo} was wrong but we could not say what was right instead.

Any way of generic speaking could be objected to. And {lo} is not

the only thing people use for lack of something better. {le} and

{loi} are also popular alternatives.

> It may turn out that, on zipfy

> gounds for example, we want to start to use {lo} in this way and a longer

> expression for old {lo} (arguably, {su'o} already works), but that is a major

> decision, not to be made casually and without comment to speak of on a wiki

> page (and so looking remarkably like a done deal).

3. The wiki page is clearly labeled as a proposal, and it is clearly

part of the BPFK work we are doing. This is a proposal to be

discussed, amended as necessary, and voted on.

> 10: Ahah! Context makes a difference; it looked like a report but it was a

> direction ({e'u} or {e'o} or {ei} or maybe something more complex). Still,

> as read by each particular teacher and applied in a particular classroom, it

> is quite particular, so {le ctuca} and {le [[or%20%7Blei%7D|or {lei}]] selctu}.

4: So even though the speaker does not have any particular

teacher in mind, you think he should use {le} because some teacher

reading it might have a particular one in mind? What about other

readers that may simply be interested in teaching methods but not

in actually performing this particular lesson?

> 12. Even if {lo} is generic in the sense set out here. To be sure, for the

> uncountables (in English), {lo djacu} comes pretty close to being about the

> substance in extension at least. But that doesn't work for countables lo

> bakni are cattle, not beef.

5:{lo tu'o gerku} is proposed for the dog all over the pavement.

"Beef" however is probably better as {bakni rectu}.

({ractu rectu} for rabbit.)

> 15: Where is the implicit negation in {nitcu}? To be sure, needing implies

> lacking; but it does not assert it.

6:Does {claxu} assert not having, or does it just imply it?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT

John E Clifford wrote:

>1. After going to meetings and reading list for 30 years, I can assure =

that — whatever they say publicaly — every person who has spent enough =

time on Loglan or Lojban to feel up to talking about it has something the=

y want to change. Some many things or broad changes, some details, almos=

t everyone additions.

>=20

>2: There have been no shortage of alternatives proposed, just none that =

got the right sort of support behind it (I think I have made three and & =

at least two others — but those cancel eachother out, of course joke).=

The problems with {le} (which is a non-starter) and {loi} (which does =

one small part of the work in some contexts) is that they, like {lo}, alr=

eady have clear uses prescribed. I think that too many people have thoug=

ht that Lojban was set in concrete or that no proposal would be accepted =

(because there were those in power who held that nothing may change — a =

holdover from Loglan, where it was often true, unless you could convice J=

CB that he thought of it, which did happen occasionally). So the slogged=

on with makeshifts rather than getting together a good case: examples, c=

lear explanation why nothing currently works, clear explanation of how th=

e proposed extension works, estimates of cost and advantage and so on. Th=

at is a lot of work for one person to do and

> loCCan has not been very good at creating committees that actually do t=

hings (what gets done gets done by one person doing it). My comment was =

an attempt to shortcut the process slightly: after fifty years of carping=

it is clear that there are something which we want to say but which all =

are attempts to say in current language have ended in failure. Let's jus=

t create ways to say them and get on with it. If we do figure out how to=

say them without all the additions (and once they get said a lot that is=

a real possibility) then we can drop the additons and retrofit the text =

corpus.

>

This isn't a bad description of where we are now. Intensionality is=20

essential, and the Book's definition of lo is a close approximation of=20

that, but unfortunately also conflated it with the extensional "da poi", =

resulting in a contradiction. Most usage of lo is intensional. Other=20

attempts at intension used bizarre stunts like lo jai ka, appropriations =

of other cmavo such as lo'e, or evasions like le. And the (only) other=20

sense of the old lo is easily expressed using su'o!

The conclusion is clear. lo must go from usually being intensional to=20

being always intensional.

We will not hammer out all the oddities of intensionality here on this=20

list before the BF must vote. The BF commissioners should vote yes=20

because this plan improves clarity and consistency, and because it's=20

better than the status quo or anything that will be sketched up before=20

the vote. But regardless of the BF's decision I will continue to apply=20

the XS in my usage as I have been. If this is a fork or schism, so be it.=

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT

xod:

> http://www.lojban.org/tiki/3Dgadri+report%2C+aug+2003

That seems broken. The address is:

http://www.lojban.org/tiki/gadri+report%2C+aug+2003

or

http://www.lojban.org/tiki/gadri+report%2C+aug+2003[[tiki-editpage.php?page=http%3A%2F%2Fwww.lojban.org%2Ftiki%2Ftiki-index.php%3Fpage%3Dgadri%2Breport%252C%2Baug%2B2003|?]]

for those of you reading from the wiki.

Thanks for reminding us of that report! Nick is undoubtedly a

better expounder than I am.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 29 of May, 2004 23:31 GMT


(lo/le ctuca cu fatri lo/le selctu mu lo vo tadni)

> 4: I don't think that a teacher reading it for guidance has anydoubt who the

> writer has in mind: the teacher reading this copy for guidance. The casual

> reader doesn't either: the teacher who is using the guidelines in an actual

> situation. This is a perhaps metaphorical sense of in mind

Yes. Is {le} supposed to be used in this metaphorical sense?

I have often been in doubt about this. Is it always for things

that are identified by the speaker, or can it be used for things

that are not identified by the speaker but which would be

identifiable by someone taking part in the (hypothetical)

situation being described?

Example:

Get two boxes of different sizes and put the smaller one

inside the bigger one.

Can we use {le} for "the bigger one" and "the smaller one" even

though there are no actual objects that the speaker has in mind?

He doesn't even know if the first part of the command will be

fulfilled, so at this point the two boxes are hypothetical.

Can he refer to them with {le}?

> 5: But number may not be irrelevant here, one is quite capable of being

> interested in reporting a smear of two-dogs on the highway. I forget all the

> proposals and all the cases for and against; I just note this has been a

> problem over the years, so let's fix it, The {rectu} only works for the

> (mainly) edible parts of critters, not for goo.

For two dog smear we can use {lo tu'o lo re gerku}.

> 6:Does {claxu} assert not having, or does it just imply it?

> 6: Asserts.

So you would say:

mi claxu roda

= mi na ponse roda

= su'oda naku zo'u mi ponse da

mi claxu su'oda

= mi na ponse su'oda

= noda zo'u mi ponse da

"I lack everything" = "There's something I don't have"?

"I lack something" = "There's nothing that I have"?

It doesn't seem right. With claxu = narponse, i.e. shortest

scope negation, it works much better.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

1. {ro} makes a lot of sense in normative discourse, where casuistry is available to deal with the hard cases, but won't do in descriptive discourse, where exceptions are not allowed (officially — but look at raw notes in real science). So {ro} is not a general solution for generics.

2: And here is the problem. Lacking casuistry (or at least without clear notions of how to apply it) {ro} won't do here. A goodly number of lions never hunt for food at night (it's easier to see it in the daylight and there is plenty around) and all of them take time off occasionally, so the exceptions are not peculiar enough for dodging an "all" (unlike the cow with the amputation who doesn't count against "all cows have four legs") The pointabout trating the various {lo}s differently may be right: certainly the meal could be {su'o cidja} (in the scope of two {lo}s), but someone might argue that nights and lions are on a par here. The objection to {so'a} at this point is that it continues the suggestion that these things are about how many critters do something, rather than being loose talk about critterkind — compare {lo'e} at least. (Notice that I am not recommending {lo} for this usage but talking about it in a context where {lo} seems best uderstood in this way.)

3: You don't have too far to find pictures worth one word ("Shit" typically) or - in the original sense of this maxim that don't help you find/do/understand anything at all ("Modern Art" (pe'a) is a case in point). But {pe'a} might be a good idea here; I think {ro} is less so.

4: the script analogy is a good one, since that is what these kinds of instruction are to a great extent. The point about other than first references is standard stuff (thjough we forget it a lot), but I think the first one is {le} too.

xod wrote:

Judging from what Nick wrote at

http://www.lojban.org/tiki/3Dgadri+report%2C+aug+2003

this is a field of active research in linguistics/philosophy, and we

should not expect to arrive at a consensus solution here. To do so would

mean that we either had broken significant ground worthy of publishing,

or more likely, that we had succeeded in deluding ourselves. The BF

commissioners should attempt to clarify the situation beyond its current

state, but not expect a solution that lies beyond all criticism. ju'a

..e'unai lo prane cu bradi lo xamgu

Jorge Llamb=EDas wrote:

>pc:

> =20

>

>>And

>>most of these are peculiar cases (if really cases of {lo} at all):

>>generalities, gnomic utterances, maxims and the like =96 things that ar=

e more

>>or less universal; that is not {lo} home ground (and very likely not it=

s

>>ground at all).

>> =20

>>

>

>But there is nothing peculiar about these sentences. They are

>everyday things people say, and which a fluent Lojban speaker

>should be able to produce without a second of hesitation. I found

>most of the English sentences with simple Google searches, I did=20

>not make any of them up myself. If such sentences cannot be produced=20

>easily with current Lojban, then there is something wrong with=20

>current Lojban.

>

>Talking about generalities is basic, it happens all the time.

>I take it you would use your proposed {xo'o} for many of the examples.

>That's a possibility. The disadvantage is that most resulting texts,

>which will certainly be full of {xo'o}s because general claims

>are very common, will not look like the Lojban that has been produced

>in the last twenty years or so. With {lo}, on the other hand, Lojban

>will continue to look like so-far-Lojban.=20

>

> =20

>

>>ei lo verba cu mutce fraxu lo makcu prenu

>>Children should always show great forbearance

>>toward grown-up people.

>> =20

>>

>It is kind of a maxim, yes. I cannot tell from your words whether=20

>you approve of this translation or whether you would translate it=20

>differently. How would you translate maxims, which are relatively=20

>frequent in any language?=20

> =20

>

1: It seems to me that maxims, and claims that sweep similarly broad,

should be translated with ro, not lo. And since ro applies to every

member, and hence to the whole type by definition, then in cooperative

usage lo will be used to discuss subsets, often but not necessarily prope=

r.

>>ku'i uinai mi na viska lo lanme pa'o lo tanxe

>>i ju'ocu'i mi milxe simsa lo makcu prenu

>>But I, alas, do not see sheep through the walls of

>>boxes. Perhaps I am a little like the grown-ups.

>>=20

>>I am not sure whether {pa'o} works like this, but the {lo}s in the firs=

t

>>sentence work out right. A good example (though perhaps for later), si=

nce it

>>reminds us that universals in negative contexts are expressed existenti=

ally:

>>=93any sheep through any box=94 (is =93the walls of=94 just a flourish?=

This eems to

>>apply as well to looking through a tubular box lacking both ends.=20

>> =20

>>

>

>The original doesn't mention walls: "Mais moi, malheureusement, je ne sa=

is

>pas voir les moutons =E0 travers les caisses." I guess context helps mak=

e it=20

>clear what is meant: The author has drawn a box, and the little prince

>is very happy with the sheep he says is inside of the box. He had reject=

ed

>all the previous attempted drawings of sheep for one reason or another.

>So at least for negative generic claims you approve of the use of lo.

>(I would take {mi na viska su'o lamne pa'o su'o tanxe} to be a more

>concrete claim, though.)

>=20

> =20

>

>>The {lo}

>>in the second sentence is probably about a species (etc.) since it is g=

oing

>>on to some property. I would use {la'e} here, but that is only a reaso=

nable

>>start of working out how to talk about species.

>> =20

>>

>

>But {la'e} still requires another gadri. Do you mean {la'e lo makcu pren=

u}?

>Is that the same as {la'e su'o makcu prenu}?=20

>

>The only use of {la'e} I know is with {la'e di'u}, to refer to what the=20

>previous sentence says. So {la'e di'u cinri} is "that's interesting", no=

t=20

>the previous sentence itself but what it says. Is that the same {la'e}=20

>that takes you from a grown-up to grown-ups in general?

>

> =20

>

>>ca lo nicte lo cinfo cu kalte lo cidja

>>At night lions hunt for food.

>> =20

>>

>

>I can't tell from your words whether you approve or not of

>this translation.

> =20

>

2:This is a quasi-definitional sentence such as we might expect to find in

an encyclopedia. Hence, I suggest {ca lo nicte ro cinfo cu kalte lo

cidja}, with possible shuffling if needed to avoid scope side effects.

Such a claim is a universal claim, not simply a non-specific one. {ca lo

nicte lo cinfo cu kalte lo cidja} should not be interpreted as general

claim about lions any more than it's a general claim about nights or

food. (It should be clear that the treatment of lion in that sentence

should be tagged differently than nights and food.) If you want to

wiggle out of making an absolute claim refuted by a single wacky lion,

so'a cinfo will do.

>>lo pa pixra cu se vamji lo ki'o valsi

>>One picture is worth a thousand words.

>>

>>Ah, I forgot this aspect of your work with quantifiers. {lo ki'o valsi=

}

>>looks OK and not noticeably different from {ki'o valsi}=20

>> =20

>>

>

>But it is noticeably different. The picture is worth the same as the

>thousand words together, it is not worth the same as a word 1000 times.=20

>{ki'o valsi} would claim that there are 1000 x which are words, such tha=

t=20

>the picture is worth x. So for example, the picture is worth "the", the=20

>picture is worth "little", the picture is worth "house", etc, 1000=20

>times. With {lo ki'o valsi} we are talking of a whole bunch of 1000=20

>words put together.=20

> =20

>

3: Here is another case for ro pixra. Use pe'a as nerd-proofing, lest some

lamer produce a picture worth only 999 words.

>>=96 presumably the

>>words could be spelled out in each case, maybe several different ways,

>>indeed. Presumably this is gnomic again so the first {lo} is either

>>universal or about species or perhaps {la=92e}.

>> =20

>>

>

>So what is a Lojban speaker to say?

>

> =20

>

>>de'i li 1960 lo pare sovda cu fepni li 42

>>In 1960 a dozen eggs cost 42 cents.

>>

>>Same old, same old. It was not just one dozen but just about any dozen=

there

>>was =96 implicit exception in force (crested floo-floo birds=92 =96 now=

extinct =96

>>eggs, certified organic, =85). Iam inclining more and more to {la=92e}=

here.

>> =20

>>

>

>Wouldn't that turn {la'e di'u} into generic "sentences like the previous=

=20

>one", instead of "the referent of the previous sentence"? That's too muc=

h

>of a change on existing usage, and besides we would need a new way of

>doing {la'e di'u}. (In fact, I think it would be great to assign say

>{tau} and {tei} to {la'e di'u} and {la'e de'u}, but that's another=20

>thread altogether.)

>

> =20

>

>>lo ctuca cu fendi lo selctu mu lo vo tadni

>>The teacher will divide the class

>>into five groups of four students.

>>

>>Hey, some basic cases, though {le ctuca} makes better sense — this se=

ems to

>>be a particular occasion. So, come to that, {le selctu} or even {lei se=

lctu}.

>> But the {mu lo vo tadni} is nice.

>> =20

>>

>

>The English sentence can be generic too, and in context it was:

>

>"LESSON SUMMARY: 30 minutes

>

>The teacher will ask the students to brainstorm ideas for each column.=20

>For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim=85)

>

>The teacher will record an answer on each piece of oak tag in the column=

..

>

>The teacher will divide the class into five groups of four students.

>

>The teacher will take all of the oak tag pieces and place them face down=

=20

>in groups according to characters, setting, and problem. The teacher wil=

l=20

>ask each group to choose one piece of oak tag from each group. ..."

>

>It is not about some particular teacher or class that the speaker=20

>has in mind. It is more general.

> =20

>

4: It could be argued that the author is writing a script and has a

particular scene in mind, and in that sense is referring to that

specific teacher. I would expect the all-but-first references to use le;

the first reference having grabbed a random teacher out of the air, and

the following references referring to that teacher and only that one.

--=20

Motorists honked in celebration in this Ramadi as news spread of the=20

assassination of the president of the Iraqi Governing Council Ezzidin Sal=

im=20

Monday. "The GC is nothing," one man shouted. "They are not the Governing=

=20

Council. They are the Prostitution Council."



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

John E Clifford wrote:

>1. {ro} makes a lot of sense in normative discourse, where casuistry is=

available to deal with the hard cases, but won't do in descriptive disco=

urse, where exceptions are not allowed (officially — but look at raw no=

tes in real science). So {ro} is not a general solution for generics.

> =20

>

If ro is insufficient, then so'a will have to do. Everything else is=20

handwaving.

>2: And here is the problem. Lacking casuistry (or at least without clea=

r notions of how to apply it) {ro} won't do here. A goodly number of lio=

ns never hunt for food at night (it's easier to see it in the daylight an=

d there is plenty around) and all of them take time off occasionally, so =

the exceptions are not peculiar enough for dodging an "all" (unlike the c=

ow with the amputation who doesn't count against "all cows have four legs=

") The pointabout trating the various {lo}s differently may be right: ce=

rtainly the meal could be {su'o cidja} (in the scope of two {lo}s), but s=

omeone might argue that nights and lions are on a par here. The objectio=

n to {so'a} at this point is that it continues the suggestion that these =

things are about how many critters do something, rather than being loose =

talk about critterkind — compare {lo'e} at least. (Notice that I am not =

recommending {lo} for this usage but talking about it in a context where =

{lo} seems best uderstood in this way.)

> =20

>

I reject the attempt to make general (all/most) claims but then slither=20

out of responsibility. If your definitional concepts don't apply to most =

of the targets, your definition needs fixing, but the gadri shouldn't=20

help you make such dishonest claims. If most cows no longer have 4 legs, =

revise your definition of "cow".

You see, making claims about the general cow is different from treating=20

cows intensionally. "Sam fears cows"; this is a claim about Sam. "Cows=20

have 4 legs" is a claim about cows. Claims about something are=20

extensional: we validate them by looping over instances and testing the=20

assertion. Intensional claims escape this, because they don't actually=20

make claims about the thing at hand. This is why Sam can fear cows in a=20

cow-less universe. Cows can appear and disappear, but the statement=20

holds invariantly true!

Perhaps we can test in/extensionality by asking: does the statement hold =

true if all X were to vanish? I would very much need doctors after all=20

doctors are slaughtered, but lions wouldn't eat at night or day if they=20

were all killed.

Have I now succeeded in ripping non-specific away from intensionality?=20

mi viska lo ractu is non-specific but hardly intensional by the above tes=

t.

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

xod:

> It seems to me that maxims, and claims that sweep similarly broad,

> should be translated with ro, not lo. And since ro applies to every

> member, and hence to the whole type by definition, then in cooperative

> usage lo will be used to discuss subsets, often but not necessarily prope=

> r.

Each and every child being forgiving of each and every grown-up

just doesn't work, as most children don't even come into

contact with most grown-ups. This is just about Mr Child being

forgiving of Mr Grown-up, or "children" being forgiving of

"grown-ups", it's not about counting instances. The maxim says

that things ought to be such that the child forgives the grown-up.

Which child? Which grown-up? How many of each? Those are the wrong

questions to ask because we are not talking about instances.

> >>ca lo nicte lo cinfo cu kalte lo cidja

> >>At night lions hunt for food.

>

> This is a quasi-definitional sentence such as we might expect to find in

> an encyclopedia. Hence, I suggest {ca lo nicte ro cinfo cu kalte lo

> cidja}, with possible shuffling if needed to avoid scope side effects.

It's not a claim about all lions. The context might be:

"Be careful, don't stray too far from the camp. At night

lions hunt for food." You are giving information about

lions and nights, but of a generic kind, not about instances.

Maybe you're lucky and just tonight there aren't any lions

around hunting for food. It's still the case that "at night

lions hunt for food" so you ought to be careful.

> Such a claim is a universal claim, not simply a non-specific one. {ca lo

> nicte lo cinfo cu kalte lo cidja} should not be interpreted as general

> claim about lions any more than it's a general claim about nights or

> food.

Agreed.

> (It should be clear that the treatment of lion in that sentence

> should be tagged differently than nights and food.) If you want to

> wiggle out of making an absolute claim refuted by a single wacky lion,

> so'a cinfo will do.

It's not meant to be a claim about how many instances of lions

do that.

> >>lo pa pixra cu se vamji lo ki'o valsi

> >>One picture is worth a thousand words.

>

> 3: Here is another case for ro pixra. Use pe'a as nerd-proofing, lest some

> lamer produce a picture worth only 999 words.

{pe'a} is ok, but even then, ro is inadequate. The idea is not that

you examine each picture and conclude that its worth is that of

a thousand words. The idea is that in general a picture gives

information that could only be conveyed by a lot of words. So

pictures are worth a lot of words, but this is not about counting

the number of pictures this applies to.

> >>lo ctuca cu fendi lo selctu mu lo vo tadni

> >>The teacher will divide the class

> >>into five groups of four students.

> >>

> It could be argued that the author is writing a script and has a

> particular scene in mind, and in that sense is referring to that

> specific teacher. I would expect the all-but-first references to use le;

> the first reference having grabbed a random teacher out of the air, and

> the following references referring to that teacher and only that one.

Kind of like English "the"...

I read it as:

TEACHER divides CLASS into 5 STUDENT-FOURSOME

The only relevant quantifier in the statement is 5 (4 is part of

a description). If there is any specificity it is not of the usual

kind, because the speaker doesn't have any particular teacher in

mind.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

Jorge Llamb=EDas wrote:

>xod:

> =20

>

>>It seems to me that maxims, and claims that sweep similarly broad,

>>should be translated with ro, not lo. And since ro applies to every

>>member, and hence to the whole type by definition, then in cooperative

>>usage lo will be used to discuss subsets, often but not necessarily pro=

pe=3D

>>r.

>> =20

>>

>

>Each and every child being forgiving of each and every grown-up

>just doesn't work, as most children don't even come into

>contact with most grown-ups.=20

>

That's irrelevant. When they do meet, the child should be forgiving.

This is not committing the child to the difficult task of seeking out

every adult. It is nonetheless a claim about every child and every adult.

>This is just about Mr Child being

>forgiving of Mr Grown-up, or "children" being forgiving of

>"grown-ups", it's not about counting instances. The maxim says

>that things ought to be such that the child forgives the grown-up.

>Which child? Which grown-up? How many of each? Those are the wrong=20

>questions to ask because we are not talking about instances.

>

> =20

>

>>>>ca lo nicte lo cinfo cu kalte lo cidja

>>>>At night lions hunt for food.

>>>> =20

>>>>

>>This is a quasi-definitional sentence such as we might expect to find i=

n

>>an encyclopedia. Hence, I suggest {ca lo nicte ro cinfo cu kalte lo

>>cidja}, with possible shuffling if needed to avoid scope side effects.

>> =20

>>

>

>It's not a claim about all lions. The context might be:

>"Be careful, don't stray too far from the camp. At night

>lions hunt for food." You are giving information about

>lions and nights, but of a generic kind, not about instances.

>Maybe you're lucky and just tonight there aren't any lions=20

>around hunting for food. It's still the case that "at night

>lions hunt for food" so you ought to be careful.

> =20

>

Consider {.iinai so'u cinfo cu kalte ca lo nicte} vs.{.iicai so'a cinfo

cu kalte ca lo nicte}. The numbers of lions hunting at night is

completely crucial. And if lions didn't exist, the warning would be

pointless — completely different from the intensional examples we cite.

>>Such a claim is a universal claim, not simply a non-specific one. {ca l=

o

>>nicte lo cinfo cu kalte lo cidja} should not be interpreted as general

>>claim about lions any more than it's a general claim about nights or

>>food.

>> =20

>>

>

>Agreed.=20

>

> =20

>

>>(It should be clear that the treatment of lion in that sentence

>>should be tagged differently than nights and food.) If you want to

>>wiggle out of making an absolute claim refuted by a single wacky lion,

>>so'a cinfo will do.

>> =20

>>

>

>It's not meant to be a claim about how many instances of lions

>do that.

>

> =20

>

>>>>lo pa pixra cu se vamji lo ki'o valsi

>>>>One picture is worth a thousand words.

>>>> =20

>>>>

>>3: Here is another case for ro pixra. Use pe'a as nerd-proofing, lest s=

ome

>>lamer produce a picture worth only 999 words.

>> =20

>>

>

>{pe'a} is ok, but even then, ro is inadequate. The idea is not that

>you examine each picture and conclude that its worth is that of

>a thousand words. The idea is that in general a picture gives=20

>information that could only be conveyed by a lot of words. So=20

>pictures are worth a lot of words, but this is not about counting

>the number of pictures this applies to.

> =20

>

Nothing here convinces me that "a picture" is not simply code for "every

picture". This is an extensional claim about every picture, regardless

of its figurative sense.

How would you interpret "A woman should wear an apron"? Doesn't it mean t=

hat we=20

loop through every woman, point to her, and demand she wear an apron?

>>>>lo ctuca cu fendi lo selctu mu lo vo tadni

>>>>The teacher will divide the class

>>>>into five groups of four students.

>>>>

>>>> =20

>>>>

>>It could be argued that the author is writing a script and has a

>>particular scene in mind, and in that sense is referring to that

>>specific teacher. I would expect the all-but-first references to use le=

>>the first reference having grabbed a random teacher out of the air, and

>>the following references referring to that teacher and only that one.

>> =20

>>

>

>Kind of like English "the"...

>

>I read it as:

>

> TEACHER divides CLASS into 5 STUDENT-FOURSOME

>

>The only relevant quantifier in the statement is 5 (4 is part of

>a description). If there is any specificity it is not of the usual

>kind, because the speaker doesn't have any particular teacher in

>mind. =20

> =20

>

What teacher? Any teacher? No, not any teacher. The teacher in the

example. Our hypothetical teacher about whom we know nothing except that

he's teaching Scooby Doo to a bunch of 3rd graders. If each reference to

a teacher referred once again to any, non-specific teacher, each

sentence might refer to a bi'u teacher, rendering the script nonsensical.

We could have given him a name, *unlike* the needed doctor.

Suppose the instructions included a second teacher. Would that be any

teacher? Again no, because "any teacher" could include the first teacher.

Furthermore, this example is totally prenexable: "Let there be a teacher.=

Let=20

there be a classroom..."

--=20

Motorists honked in celebration in this Ramadi as news spread of the=20

assassination of the president of the Iraqi Governing Council Ezzidin Sal=

im=20

Monday. "The GC is nothing," one man shouted. "They are not the Governing=

=20

Council. They are the Prostitution Council."



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

Jorge Llamb=EDas wrote:

>xod:

> =20

>

>>It seems to me that maxims, and claims that sweep similarly broad,

>>should be translated with ro, not lo. And since ro applies to every

>>member, and hence to the whole type by definition, then in cooperative

>>usage lo will be used to discuss subsets, often but not necessarily pro=

pe=3D

>>r.

>> =20

>>

>

>Each and every child being forgiving of each and every grown-up

>just doesn't work, as most children don't even come into

>contact with most grown-ups. This is just about Mr Child being

>forgiving of Mr Grown-up, or "children" being forgiving of

>"grown-ups", it's not about counting instances.

>

I don't think we can simply say "don't worry about instances" without=20

proving that counting instances is absurd in this situation. Little=20

Johnny should forgive Aunt Susan; those are specific instances and it=20

makes total sense. The maxim is only a numeric generalization from this=20

extensional case.

In the situation of needed doctors, counting doctor instances is=20

demonstrably absurd.

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT


> That's irrelevant. When they do meet, the child should be forgiving.

> This is not committing the child to the difficult task of seeking out

> every adult. It is nonetheless a claim about every child and every adult.

To me {ei ro verba cu fraxu ro makcu prenu} means:

"It ought to be the case that each child forgives each adult."

What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

> Consider {.iinai so'u cinfo cu kalte ca lo nicte} vs.{.iicai so'a cinfo

> cu kalte ca lo nicte}. The numbers of lions hunting at night is

> completely crucial.

In those claims, indeed it is. But in {ca lo nicte lo cinfo cu kalte

lo cidja} no numbers are mentioned.

> And if lions didn't exist, the warning would be

> pointless — completely different from the intensional examples we cite.

The fact that you can quantify does not mean that you must.

If you want to be precise with tense, you can be. But Lojban does

not force it upon you. If you want to be precise with number, you

can be. But Lojban does not force it upon you. A claim with the

minimal gadri {lo} cannot be false on account of number, because

it doesn't carry any info on number.

> >>>>lo pa pixra cu se vamji lo ki'o valsi

> >>>>One picture is worth a thousand words.

>

> Nothing here convinces me that "a picture" is not simply code for "every

> picture". This is an extensional claim about every picture, regardless

> of its figurative sense.

If you want to make the extensional claim, nothing stops you,

but it has a different sense altogether. You'd be saying that

every picture is very informative instead of comparing the

informative value of pictures vs. words.

> How would you interpret "A woman should wear an apron"? Doesn't it mean t=

> hat we=20

> loop through every woman, point to her, and demand she wear an apron?

It is very similar to "Every woman should wear an apron", yes.

In Lojban you can be as precise or imprecise as you like.

>From least to most precise:

ei lo ninmu cu dasni lo cragai

ei ro ninmu cu dasni lo cragai

ei ro ninmu cu dasni su'o cragai

ei ro ninmu cu dasni pa cragai

And of course you can add tense:

ei ro ninmu ca ca'o ca'a dasni pa cragai

"Every woman should at this moment be actually wearing an apron."

If you don't express the time some of them might argue that they

wore one yesterday and so they already fulfilled their duty.

> >>>>lo ctuca cu fendi lo selctu mu lo vo tadni

> >>>>The teacher will divide the class

> >>>>into five groups of four students.

>

> What teacher? Any teacher? No, not any teacher. The teacher in the

> example. Our hypothetical teacher about whom we know nothing except that

> he's teaching Scooby Doo to a bunch of 3rd graders.

And who might never exist, right.

> If each reference to

> a teacher referred once again to any, non-specific teacher, each

> sentence might refer to a bi'u teacher, rendering the script nonsensical.

> We could have given him a name, *unlike* the needed doctor.

Not nonsensical, just more vague. But context will help sort it out.

When telling a story we don't need to put a tense in every sentence,

as usually things are told in the order they happened. You can, of

course be more precise when you need or want to.

> Suppose the instructions included a second teacher. Would that be any

> teacher? Again no, because "any teacher" could include the first teacher.

If it's important that it's about two teachers then you'd have to use

number, of course. If it's irrelevant if there is one teacher or two

conducting the lesson, you might not even mention number.

> Furthermore, this example is totally prenexable: "Let there be a teacher.=

> Let=20

> there be a classroom..."

Indeed, and there's nothing wrong with doing it that way.

But let's get away from the idea that there is always one

correct gadri for each situation and all the rest are wrong.

{lo} is the most general gadri and so it will practically

never be wrong in cases when another gadri is more precise.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 02:44 GMT

xod:

> I don't think we can simply say "don't worry about instances" without=20

> proving that counting instances is absurd in this situation. Little=20

> Johnny should forgive Aunt Susan; those are specific instances and it=20

> makes total sense. The maxim is only a numeric generalization from this=20

> extensional case.

>

> In the situation of needed doctors, counting doctor instances is=20

> demonstrably absurd.

But lo is not restricted to cases where counting is absurd! It is

for cases where counting is irrelevant, be it because it is absurd

or because it is just not important.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 19:16 GMT

A: Phooey!. The Books description of {lo} has nothing to do with intensionality and indeed the idea of an intensional gadri barely makes sense: it could not be used in transparent context and would be unnecessary in opaque one. What {lo} seems to be used all to often to shoot for is that wondrously vague sense of English plurals, "the" generic expressions and the like, which are closer to {lo'e} than anything else presently in the language: that is they talk about the members of a class but without an specific number being relevant — one is usually too few, all is usually more than is strictly required, and within that not all count equally for the claim. This is not intensional but just a different way of looking at a class, by weight, as it were, rather than by number. It is also different from thinking about the class itselr as a node in the conceptual tree — closer to {lo'i} but again more somewhat isolated from the members (though {lo'i} may work jhere — my proposal just

leaves the answer to that for later and meanwhile gets on with business). To be sure, the difference between a set and the property that defines its are sometimes said to be the differnce between extension and intension, but that is a different distinction by that name from the corresponding talk about contexts (though there may be some deep or remote connection).

B: Well, we ought to find some way of expreessing generic usage, but that it be {lo} is at least controversial. that {lo} has been misused (against the Book) in this way in the past hardly justifies continuing to do it.

John E Clifford wrote:

>1. After going to meetings and reading list for 30 years, I can assure =

that — whatever they say publicaly — every person who has spent enough =

time on Loglan or Lojban to feel up to talking about it has something the=

y want to change. Some many things or broad changes, some details, almos=

t everyone additions.

>=20

>2: There have been no shortage of alternatives proposed, just none that =

got the right sort of support behind it (I think I have made three and & =

at least two others — but those cancel eachother out, of course joke).=

The problems with {le} (which is a non-starter) and {loi} (which does =

one small part of the work in some contexts) is that they, like {lo}, alr=

eady have clear uses prescribed. I think that too many people have thoug=

ht that Lojban was set in concrete or that no proposal would be accepted =

(because there were those in power who held that nothing may change — a =

holdover from Loglan, where it was often true, unless you could convice J=

CB that he thought of it, which did happen occasionally). So the slogged=

on with makeshifts rather than getting together a good case: examples, c=

lear explanation why nothing currently works, clear explanation of how th=

e proposed extension works, estimates of cost and advantage and so on. Th=

at is a lot of work for one person to do and

> loCCan has not been very good at creating committees that actually do t=

hings (what gets done gets done by one person doing it). My comment was =

an attempt to shortcut the process slightly: after fifty years of carping=

it is clear that there are something which we want to say but which all =

are attempts to say in current language have ended in failure. Let's jus=

t create ways to say them and get on with it. If we do figure out how to=

say them without all the additions (and once they get said a lot that is=

a real possibility) then we can drop the additons and retrofit the text =

corpus.

>

A: This isn't a bad description of where we are now. Intensionality is=20

essential, and the Book's definition of lo is a close approximation of=20

that, but unfortunately also conflated it with the extensional "da poi", =

resulting in a contradiction. Most usage of lo is intensional. Other=20

attempts at intension used bizarre stunts like lo jai ka, appropriations =

of other cmavo such as lo'e, or evasions like le. And the (only) other=20

sense of the old lo is easily expressed using su'o!

B:e conclusion is clear. lo must go from usually being intensional to=20

being always intensional.

We will not hammer out all the oddities of intensionality here on this=20

list before the BF must vote. The BF commissioners should vote yes=20

because this plan improves clarity and consistency, and because it's=20

better than the status quo or anything that will be sketched up before=20

the vote. But regardless of the BF's decision I will continue to apply=20

the XS in my usage as I have been. If this is a fork or schism, so be it.=

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 30 of May, 2004 19:16 GMT

A: Actually, "don't worry about instances" is more or less a rule in normative discouse. At least the apparent counterinstances are dismissed with some small argument — ax murderers are adults to be forgiven by any one. Not counting doctors in the "need" case is just a feature of opaque context: the relevant doctors are not arounf to be counted. Maxims, however are not numeric generalizations — if they are generalizations; they are weighted, generic, rules about classes (or perhaps laying obs on such genric claims about classes).

xod wrote:Jorge Llamb=EDas wrote:

>xod:

> =20

>

>>It seems to me that maxims, and claims that sweep similarly broad,

>>should be translated with ro, not lo. And since ro applies to every

>>member, and hence to the whole type by definition, then in cooperative

>>usage lo will be used to discuss subsets, often but not necessarily pro=

pe=3D

>>r.

>> =20

>>

>

>Each and every child being forgiving of each and every grown-up

>just doesn't work, as most children don't even come into

>contact with most grown-ups. This is just about Mr Child being

>forgiving of Mr Grown-up, or "children" being forgiving of

>"grown-ups", it's not about counting instances.

>

A:I don't think we can simply say "don't worry about instances" without=20

proving that counting instances is absurd in this situation. Little=20

Johnny should forgive Aunt Susan; those are specific instances and it=20

makes total sense. The maxim is only a numeric generalization from this=20

extensional case.

In the situation of needed doctors, counting doctor instances is=20

demonstrably absurd.

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 30 of May, 2004 19:17 GMT

A: Well, of course no child (or anyone else) has to forgive someone he never heard of, nor (more significantly) are they bound to forgive the unforgivable. On the first issue, it seems fair to understand the claim (as I did-- misrecalling the actual example — when I said that old {lo} seemed right for the case) as "if a child (x) is crossed by an adult (y), x should forgive y" The protasis here is lost in conversational implicature (presupposition), leaving only a stripped down version of the apodasis, where the {lo}s ought logically be connverted to {ro}s. But you can't depend on grammar to be logical all the time, even in Lojban. In this case, the maxim can be taken as obligating a generic claim — which is probablty not true, else why oblicate it — to the effect that children forgive adults. I think that the {ro} reading is more comfortable to most moralists, but the generic one is not impossible.

B: Aside from the issue of whether {lo} is the right word here (and whether this is the way you have been using {lo}), this eems right for generic usage.

Jorge Llambías wrote:


> That's irrelevant. When they do meet, the child should be forgiving.

> This is not committing the child to the difficult task of seeking out

> every adult. It is nonetheless a claim about every child and every adult.

A:To me {ei ro verba cu fraxu ro makcu prenu} means:

"It ought to be the case that each child forgives each adult."

What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

> Consider {.iinai so'u cinfo cu kalte ca lo nicte} vs.{.iicai so'a cinfo

> cu kalte ca lo nicte}. The numbers of lions hunting at night is

> completely crucial.

In those claims, indeed it is. But in {ca lo nicte lo cinfo cu kalte

lo cidja} no numbers are mentioned.

> And if lions didn't exist, the warning would be

> pointless — completely different from the intensional examples we cite.

B:The fact that you can quantify does not mean that you must.

If you want to be precise with tense, you can be. But Lojban does

not force it upon you. If you want to be precise with number, you

can be. But Lojban does not force it upon you. A claim with the

minimal gadri {lo} cannot be false on account of number, because

it doesn't carry any info on number.

> >>>>lo pa pixra cu se vamji lo ki'o valsi

> >>>>One picture is worth a thousand words.

>

> Nothing here convinces me that "a picture" is not simply code for "every

> picture". This is an extensional claim about every picture, regardless

> of its figurative sense.

If you want to make the extensional claim, nothing stops you,

but it has a different sense altogether. You'd be saying that

every picture is very informative instead of comparing the

informative value of pictures vs. words.

> How would you interpret "A woman should wear an apron"? Doesn't it mean t=

> hat we=20

> loop through every woman, point to her, and demand she wear an apron?

It is very similar to "Every woman should wear an apron", yes.

In Lojban you can be as precise or imprecise as you like.

>From least to most precise:

ei lo ninmu cu dasni lo cragai

ei ro ninmu cu dasni lo cragai

ei ro ninmu cu dasni su'o cragai

ei ro ninmu cu dasni pa cragai

And of course you can add tense:

ei ro ninmu ca ca'o ca'a dasni pa cragai

"Every woman should at this moment be actually wearing an apron."

If you don't express the time some of them might argue that they

wore one yesterday and so they already fulfilled their duty.

> >>>>lo ctuca cu fendi lo selctu mu lo vo tadni

> >>>>The teacher will divide the class

> >>>>into five groups of four students.

>

> What teacher? Any teacher? No, not any teacher. The teacher in the

> example. Our hypothetical teacher about whom we know nothing except that

> he's teaching Scooby Doo to a bunch of 3rd graders.

And who might never exist, right.

> If each reference to

> a teacher referred once again to any, non-specific teacher, each

> sentence might refer to a bi'u teacher, rendering the script nonsensical.

> We could have given him a name, *unlike* the needed doctor.

Not nonsensical, just more vague. But context will help sort it out.

When telling a story we don't need to put a tense in every sentence,

as usually things are told in the order they happened. You can, of

course be more precise when you need or want to.

> Suppose the instructions included a second teacher. Would that be any

> teacher? Again no, because "any teacher" could include the first teacher.

If it's important that it's about two teachers then you'd have to use

number, of course. If it's irrelevant if there is one teacher or two

conducting the lesson, you might not even mention number.

> Furthermore, this example is totally prenexable: "Let there be a teacher.=

> Let=20

> there be a classroom..."

Indeed, and there's nothing wrong with doing it that way.

But let's get away from the idea that there is always one

correct gadri for each situation and all the rest are wrong.

{lo} is the most general gadri and so it will practically

never be wrong in cases when another gadri is more precise.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 19:17 GMT

pc:

> What {lo} seems to be used all to often to shoot for is that

> wondrously vague sense of English plurals,

English bare plurals, yes. Not "some brodas" or "all brodas" or

"the brodas", but just "brodas".

>"the" generic expressions and the

> like, which are closer to {lo'e} than anything else presently in the

> language:

That's why I used {lo'e} for that for many years, but it didn't

catch on. And even to me, it always seemed too marked, and there

was always someone around to point out that that's not what

"the typical" means.

> that is they talk about the members of a class but without an

> specific number being relevant — one is usually too few, all is usually more

> than is strictly required, and within that not all count equally for the

> claim. This is not intensional but just a different way of looking at a

> class, by weight, as it were, rather than by number. It is also different

> from thinking about the class itselr as a node in the conceptual tree --

> closer to {lo'i} but again more somewhat isolated from the members (though

> {lo'i} may work jhere — my proposal just

> leaves the answer to that for later and meanwhile gets on with business).

Your proposal at this point is to use {lo'e}, right?

> To be sure, the difference between a set and the property that defines its

> are sometimes said to be the differnce between extension and intension, but

> that is a different distinction by that name from the corresponding talk

> about contexts (though there may be some deep or remote connection).

The difference in terms of sets that I'm familiar with is in how

a set is defined. A definition by extension is a list of the

members, whereas a definition by intension is giving the property

that the members have. So the same set A can be defined either way:

by extension A={2,4,6,8}

by intension A={x/x is an even number greater than 1 and less than 9}

The same set can have different definitions by intension.

Lojban uses {ce} for definitions by extension and {lo'i} for definitions

by intension.

> B: Well, we ought to find some way of expreessing generic usage, but that it

> be {lo} is at least controversial. that {lo} has been misused (against the

> Book) in this way in the past hardly justifies continuing to do it.

I agree that that is not in itself a justification but just a supporting

argument. Another supporting argument is that nothing is lost in terms of

expressiveness because {su'o} duplicates the job of old-lo. Also, because

the proposed sense is more general that the old and covers it, past usage

is hardly invalidated but at most may read as a little more vague than

intended. And since in a sense {lo} is supposed to be the least marked

gadri, it should go to the least restricted notion.

How do languages without articles handle this? Always using {lo}

and making distinctions of specificity by other means would be

like having a language without articles.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 30 of May, 2004 20:52 GMT

Jorge Llamb=EDas wrote:

>--- xod wrote:

> =20

>

>>That's irrelevant. When they do meet, the child should be forgiving.

>>This is not committing the child to the difficult task of seeking out

>>every adult. It is nonetheless a claim about every child and every adul=

t.

>> =20

>>

>

>To me {ei ro verba cu fraxu ro makcu prenu} means:

>"It ought to be the case that each child forgives each adult." =20

>

>What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

> =20

>

I must interpret it extensionally, resulting to "some children...some=20

adults", unchanged from the old-lo, and not a really as a maxim=20

  • because* of it's lack of ro. And the hypotheticality given by .ei might=20

be enough to avoid nerdy criticism of "all". I used to think that the=20

non-specificity of lo forced the statement to apply to the entire type,=20

but now I don't.

>=20

> =20

>

>>Consider {.iinai so'u cinfo cu kalte ca lo nicte} vs.{.iicai so'a cinfo

>>cu kalte ca lo nicte}. The numbers of lions hunting at night is

>>completely crucial.=20

>> =20

>>

>

>In those claims, indeed it is. But in {ca lo nicte lo cinfo cu kalte=20

>lo cidja} no numbers are mentioned.

> =20

>

"Some lions hunt at night" means at least one does, and only means that=20

we can no longer say that none do. It doesn't tell us about the habits=20

of lions, which I cannot see as anything but a ro statement (if part of=20

the definition of lion) or a so'a statement (if an observed property,=20

god forbid you should neglect the outliers).

>>And if lions didn't exist, the warning would be

>>pointless — completely different from the intensional examples we cite=

..

>> =20

>>

>

>The fact that you can quantify does not mean that you must.

>If you want to be precise with tense, you can be. But Lojban does

>not force it upon you. If you want to be precise with number, you=20

>can be. But Lojban does not force it upon you. A claim with the=20

>minimal gadri {lo} cannot be false on account of number, because

>it doesn't carry any info on number.

> =20

>

Other than the trivial case of zero, I agree.

>>>>>>lo pa pixra cu se vamji lo ki'o valsi

>>>>>>One picture is worth a thousand words.

>>>>>> =20

>>>>>>

>>Nothing here convinces me that "a picture" is not simply code for "ever=

y

>>picture". This is an extensional claim about every picture, regardless

>>of its figurative sense.

>> =20

>>

>

>If you want to make the extensional claim, nothing stops you,

>but it has a different sense altogether. You'd be saying that

>every picture is very informative instead of comparing the

>informative value of pictures vs. words.

> =20

>

What's the difference? The informative value of pictures vs. words is=20

exactly a word:picture mapping.

>>How would you interpret "A woman should wear an apron"? Doesn't it mean=

t=3D

>>hat we=3D20

>>loop through every woman, point to her, and demand she wear an apron?

>> =20

>>

>

>It is very similar to "Every woman should wear an apron", yes.

>In Lojban you can be as precise or imprecise as you like.

>From least to most precise:

>

> ei lo ninmu cu dasni lo cragai

> =20

>

Here is where the meaning really changes. Above is a very weak claim,=20

below is much stronger. I won't assume that lo is ro, although it's not=20

ruled out. But this is normative usage, and there's a big difference=20

between a maxim the speaker feels should apply to the type, and an=20

observational statement which may or may not generalize to the type (and=20

thus lo equal ro).

> ei ro ninmu cu dasni lo cragai

> ei ro ninmu cu dasni su'o cragai

> ei ro ninmu cu dasni pa cragai

>

>And of course you can add tense:

>

> ei ro ninmu ca ca'o ca'a dasni pa cragai

>

>"Every woman should at this moment be actually wearing an apron."=20

>If you don't express the time some of them might argue that they

>wore one yesterday and so they already fulfilled their duty.

>

> =20

>

>>>>>>lo ctuca cu fendi lo selctu mu lo vo tadni

>>>>>>The teacher will divide the class

>>>>>>into five groups of four students.

>>>>>> =20

>>>>>>

>>What teacher? Any teacher? No, not any teacher. The teacher in the

>>example. Our hypothetical teacher about whom we know nothing except tha=

t

>>he's teaching Scooby Doo to a bunch of 3rd graders.=20

>> =20

>>

>

>And who might never exist, right.

> =20

>

He exists like a fictional story character. But I think we understand=20

each other here.

>>If each reference to

>>a teacher referred once again to any, non-specific teacher, each

>>sentence might refer to a bi'u teacher, rendering the script nonsensica=

l.

>>We could have given him a name, *unlike* the needed doctor.

>> =20

>>

>

>Not nonsensical, just more vague. But context will help sort it out.

>When telling a story we don't need to put a tense in every sentence,

>as usually things are told in the order they happened. You can, of

>course be more precise when you need or want to.

>

> =20

>

>>Suppose the instructions included a second teacher. Would that be any

>>teacher? Again no, because "any teacher" could include the first teache=

r.

>> =20

>>

>

>If it's important that it's about two teachers then you'd have to use

>number, of course. If it's irrelevant if there is one teacher or two

>conducting the lesson, you might not even mention number.

>=20

> =20

>

>>Furthermore, this example is totally prenexable: "Let there be a teache=

r.=3D

>> Let=3D20

>>there be a classroom..."

>> =20

>>

>

>Indeed, and there's nothing wrong with doing it that way.

>But let's get away from the idea that there is always one

>correct gadri for each situation and all the rest are wrong.

>{lo} is the most general gadri and so it will practically=20

>never be wrong in cases when another gadri is more precise.

>

>mu'o mi'e xorxes

>

>

>

>=09

> =09

>'__

>Do you Yahoo!?

>Friends. Fun. Try the all-new Yahoo! Messenger.

>http://messenger.yahoo.com/=20

>

>

> =20

>

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 30 of May, 2004 22:02 GMT


> Jorge Llamb=EDas wrote:

> >What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

>

> I must interpret it extensionally, resulting to "some children...some=20

> adults", unchanged from the old-lo, and not a really as a maxim=20

> *because* of it's lack of ro.

Ok, but that's not the proposed lo. The proposed lo does not have

a hidden quantifier that you must glork from context. All the

sentence says, with the proposed lo, is:

"It should be so that: CHILDREN forgive ADULTS"

Nothing more. That's all the statement says. How you take that to

particular instances is up to you, it is not contained in the

sentence. It says nothing about how many children should do what

to how many adults. If you want, you can add precision in different

ways. One way is to add universal quantifiers to one or both

terms. Another way is to add tense for example:

ei roroiku fe'eroroiku lo verba cu fraxu lo makcu prenu

Everytime and everywhere children should forgive adults.

> And the hypotheticality given by .ei might=20

> be enough to avoid nerdy criticism of "all". I used to think that the=20

> non-specificity of lo forced the statement to apply to the entire type,=20

> but now I don't.

The non-specificity of lo doesn't force anything about instances.

It is simply not a statement about instances.

> "Some lions hunt at night" means at least one does, and only means that=20

> we can no longer say that none do.

Of course, that's what {su'o cinfo cu kalte ca lo nicte} means.

> It doesn't tell us about the habits=20

> of lions, which I cannot see as anything but a ro statement (if part of=20

> the definition of lion) or a so'a statement (if an observed property,=20

> god forbid you should neglect the outliers).

But {lo cinfo cu kalte ca lo nicte} is meant neither as "some lions

hunt at night" nor as "all lions hunt at night". It is meant as the

more vague "lions hunt at night", without reference to the number of

instances. If you are told that, you may or may not be interested

in enquiring further, "do they hunt only at night?" "do they hunt

every night?" "does every lion hunt at night?" "does every lion

hunt every night?" "do only lions hunt at night?" and many other

questions, none of which are answered by the original claim, though

they may be suggested by the context.

The information of number of instances is just not contained in lo,

which is good because sometimes we are not interested in it. If we

make lo overprecise, we don't have a way of expresing ourselves

when precision is not possible or not wanted.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 31 of May, 2004 05:25 GMT

Jorge Llamb=EDas wrote:

>--- xod wrote:

> =20

>

>>Jorge Llamb=3DEDas wrote:

>> =20

>>

>>>What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

>>> =20

>>>

>>I must interpret it extensionally, resulting to "some children...some=3D=

20

>>adults", unchanged from the old-lo, and not a really as a maxim=3D20

>>*because* of it's lack of ro.=20

>> =20

>>

>

>Ok, but that's not the proposed lo. The proposed lo does not have=20

>a hidden quantifier that you must glork from context. All the

>sentence says, with the proposed lo, is:

>

>"It should be so that: CHILDREN forgive ADULTS"

>

>Nothing more. That's all the statement says. How you take that to

>particular instances is up to you, it is not contained in the=20

>sentence.=20

>

How can it not be there from context? Part of the context of lo is that

the speaker chose to use lo instead of some other gadri. Given that

maxims are supposed to apply universally, and given that the speaker

avoided ro, and given that the speaker is being cooperative, we must

conclude that it's not a maxim. I am not saying that lo in this case is

incorrect, but that it is unhelpful.

>It says nothing about how many children should do what

>to how many adults. If you want, you can add precision in different

>ways. One way is to add universal quantifiers to one or both

>terms. Another way is to add tense for example:

>

> ei roroiku fe'eroroiku lo verba cu fraxu lo makcu prenu

> Everytime and everywhere children should forgive adults.

>

> =20

>

>>And the hypotheticality given by .ei might=3D20

>>be enough to avoid nerdy criticism of "all". I used to think that the=3D=

20

>>non-specificity of lo forced the statement to apply to the entire type,=

=3D20

>>but now I don't.

>> =20

>>

>

>The non-specificity of lo doesn't force anything about instances.

>It is simply not a statement about instances.

> =20

>

I have a hard time not reducing lo to an extensional claim in an

extensional context such as this. Particularly, when the statement

really should be derived from observations and falsifiable by other

observations. Surely you agree that if NO lions hunt at night, then this

non-extensional claim would be false, yes?

--=20

Motorists honked in celebration in this Ramadi as news spread of the=20

assassination of the president of the Iraqi Governing Council Ezzidin Sal=

im=20

Monday. "The GC is nothing," one man shouted. "They are not the Governing=

=20

Council. They are the Prostitution Council."



Posted by Anonymous on Mon 31 of May, 2004 19:11 GMT

xod:

> Jorge Llamb=EDas wrote:

> >"It should be so that: CHILDREN forgive ADULTS"

> >

> >Nothing more. That's all the statement says. How you take that to

> >particular instances is up to you, it is not contained in the=20

> >sentence.=20

>

> How can it not be there from context? Part of the context of lo is that

> the speaker chose to use lo instead of some other gadri.

That's not how I see it. You cannot help but to use a gadri.

Otherwise you can't make a sumti. All that {lo} does is turn

a selbri into a sumti, it adds nothing else. If you want to

add precision, you have other gadri or quantifiers for that,

but using {lo} is like using {cu}, it's vacuous. You don't

choose it, it's imposed by the grammar as the minimal

selbri-to-sumti converter.

>Given that

> maxims are supposed to apply universally, and given that the speaker

> avoided ro, and given that the speaker is being cooperative, we must

> conclude that it's not a maxim. I am not saying that lo in this case is

> incorrect, but that it is unhelpful.

The speaker did not avoid {ro}, just as when you use {cu}

instead of {ca} or {pu} you are not avoiding {ca} or {pu}.

You simply don't care to be that precise.

> >The non-specificity of lo doesn't force anything about instances.

> >It is simply not a statement about instances.

>

> I have a hard time not reducing lo to an extensional claim in an

> extensional context such as this. Particularly, when the statement

> really should be derived from observations and falsifiable by other

> observations. Surely you agree that if NO lions hunt at night, then

> this non-extensional claim would be false, yes?

Yes. If you know that {lo cinfo cu kalte ca lo nicte} you

can conclude that {su'o cinfo cu kalte ca su'o nicte}.

And you can also conclude that {lo cinfo su'oroi kalte ca

lo nicte}.

But both statements with {su'o} are more precise than the

general statement without su'o.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 31 of May, 2004 19:11 GMT

1. sometimes "the brodas" too, though that seems more likely to be about species.

2. Yes, the gehavior may be common (even) without being typical and it is the common that one wants. Unlike old {lo}, which is about particular — though unspecified — individuals on particular occasions where they could be dientified, what is wanted is groups of perhaps different individuals in different situations, no longer recoverable individually. Yet factual, not definitional nor (usually) universal (which get closer to species use).

3. No, a new gadri with a grammar (and a semantics) like {lo'e}, but without the stigma of typicality.

4. Yes, this is a thrid (or are we up to fourth) distinction tht is sometimes called extension-intension. It is related to the one I was working on below in that the specification in one *lists* the extensions and in the other the intension; but both listing give the set, an extension, not the property, an intension.

5. As you know, I am not convinced that old {lo} and {su'o} really are the same. Certainly there are places where {lo} occurs that {su'o} cannot in the same meaning (before internal quantifiers, for example) but they seem trivial. Nor is it clear that generic usage is more unmarked than particular, but that is a fault of the ambiguity of the notion of marking. Still, what is essential here is that we need both somehow and that we don't have them now..Since {lo} is fairly well-defined in one use, it seems natural to add another expression for the other use. On the other hand, it may turn out on, say, Zipfean grounds, that the generic use is so markedly more common than the particular that giving it a longer form is just criminal. Then the misuse — especially if it is buried in a lot of otherwise valuable text — might justify the change.

6. Variously, I gather. Perhaps a Russian or a Chinese expert can help here. As far as I can make out for Medieval Latin (which, admittedly was moving toward articles), they mostly did not make the distinction overtly but worked implicitly by context and explicitly by correcction when errors occurred (see the whole discussion on the proprietates terminorum which are large sorting these things out). Classical Sanskrit seems about the same and scholastic Sanskrit basically never talks about particular cases — except through lengthy periphrasis; "locus of brodanessness" or so.

Jorge Llambías wrote:

pc:

> What {lo} seems to be used all to often to shoot for is that

> wondrously vague sense of English plurals,

1.English bare plurals, yes. Not "some brodas" or "all brodas" or

"the brodas", but just "brodas".

>"the" generic expressions and the

> like, which are closer to {lo'e} than anything else presently in the

> language:

2>That's why I used {lo'e} for that for many years, but it didn't

c.tch on. And even to me, it always seemed too marked, and there

was always someone around to point out that that's not what

"the typical" means.

> that is they talk about the members of a class but without an

> specific number being relevant — one is usually too few, all is usually more

> than is strictly required, and within that not all count equally for the

> claim. This is not intensional but just a different way of looking at a

> class, by weight, as it were, rather than by number. It is also different

> from thinking about the class itselr as a node in the conceptual tree --

> closer to {lo'i} but again more somewhat isolated from the members (though

> {lo'i} may work jhere — my proposal just

> leaves the answer to that for later and meanwhile gets on with business).

3.Your proposal at this point is to use {lo'e}, right?

> To be sure, the difference between a set and the property that defines its

> are sometimes said to be the differnce between extension and intension, but

> that is a different distinction by that name from the corresponding talk

> about contexts (though there may be some deep or remote connection).

4.The difference in terms of sets that I'm familiar with is in how

a set is defined. A definition by extension is a list of the

members, whereas a definition by intension is giving the property

that the members have. So the same set A can be defined either way:

by extension A={2,4,6,8}

by intension A={x/x is an even number greater than 1 and less than 9}

The same set can have different definitions by intension.

Lojban uses {ce} for definitions by extension and {lo'i} for definitions

by intension.

> B: Well, we ought to find some way of expreessing generic usage, but that it

> be {lo} is at least controversial. that {lo} has been misused (against the

> Book) in this way in the past hardly justifies continuing to do it.

5. I agree that that is not in itself a justification but just a supporting

argument. Another supporting argument is that nothing is lost in terms of

expressiveness because {su'o} duplicates the job of old-lo. Also, because

the proposed sense is more general that the old and covers it, past usage

is hardly invalidated but at most may read as a little more vague than

intended. And since in a sense {lo} is supposed to be the least marked

gadri, it should go to the least restricted notion.

6.How do languages without articles handle this? Always using {lo}

and making distinctions of specificity by other means would be

like having a language without articles.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 31 of May, 2004 19:12 GMT

1. I think the mess comes out of its deep history, whihc may well have started with (the roots of) "if a child (x) is harmed by an adult (y), x ought to forgive y. This would become a universal conditional by one line of transformations, but the present sentence but another, which puts the condition into presupposition (how can anyone forgive anyone if not harmed by them?) aand carry the forms forward from the old condition to their consequent-anaphor. Messy, but plausible, since it works and gives the right result: this really is a maxim (we could, for comfort, get the {ro}'s by much the same process).

2. Old {lo} certainly does not work for generics (the above case is odd because of the {ei}) but {ro}, taken literally won't do either, since it is false and something true is intended. {so'a}, while literally true, misses something of the force intended. I think the idea of counting rather than weighing is probably bound to miss the point.

3. With some new {lo} (not necessarily any now flying around).

4.?? I take it that the chestnut says that pictures tell us more than words, a lot more than any single word (though probably not more than a text which uses an equal number of bytes). It is not meant to be exact since (if for no other reason) words themselves have different values, and indeed their values change depending on the other words they are with and how arranged. It does not appear to say anything about whether pictures are very informative or not, unless it is assumed that 1000 words is a lot of information.

xod wrote:

Jorge Llamb=EDas wrote:

>--- xod wrote:

> =20

>

>>That's irrelevant. When they do meet, the child should be forgiving.

>>This is not committing the child to the difficult task of seeking out

>>every adult. It is nonetheless a claim about every child and every adul=

t.

>> =20

>>

>

>To me {ei ro verba cu fraxu ro makcu prenu} means:

>"It ought to be the case that each child forgives each adult." =20

>

>What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

> =20

>

1.I must interpret it extensionally, resulting to "some children...some=20

adults", unchanged from the old-lo, and not a really as a maxim=20

  • because* of it's lack of ro. And the hypotheticality given by .ei might=20

be enough to avoid nerdy criticism of "all". I used to think that the=20

non-specificity of lo forced the statement to apply to the entire type,=20

but now I don't.

>=20

> =20

>

>>Consider {.iinai so'u cinfo cu kalte ca lo nicte} vs.{.iicai so'a cinfo

>>cu kalte ca lo nicte}. The numbers of lions hunting at night is

>>completely crucial.=20

>> =20

>>

>

>In those claims, indeed it is. But in {ca lo nicte lo cinfo cu kalte=20

>lo cidja} no numbers are mentioned.

> =20

>

2."Some lions hunt at night" means at least one does, and only means that=20

we can no longer say that none do. It doesn't tell us about the habits=20

of lions, which I cannot see as anything but a ro statement (if part of=20

the definition of lion) or a so'a statement (if an observed property,=20

god forbid you should neglect the outliers).

>>And if lions didn't exist, the warning would be

>>pointless — completely different from the intensional examples we cite=

..

>> =20

>>

>

>The fact that you can quantify does not mean that you must.

>If you want to be precise with tense, you can be. But Lojban does

>not force it upon you. If you want to be precise with number, you=20

>can be. But Lojban does not force it upon you. A claim with the=20

>minimal gadri {lo} cannot be false on account of number, because

>it doesn't carry any info on number.

> =20

>

3.Other than the trivial case of zero, I agree.

>>>>>>lo pa pixra cu se vamji lo ki'o valsi

>>>>>>One picture is worth a thousand words.

>>>>>> =20

>>>>>>

>>Nothing here convinces me that "a picture" is not simply code for "ever=

y

>>picture". This is an extensional claim about every picture, regardless

>>of its figurative sense.

>> =20

>>

>

>If you want to make the extensional claim, nothing stops you,

>but it has a different sense altogether. You'd be saying that

>every picture is very informative instead of comparing the

>informative value of pictures vs. words.

> =20

>

4. What's the difference? The informative value of pictures vs. words is=20

exactly a word:picture mapping.

>>How would you interpret "A woman should wear an apron"? Doesn't it mean=

t=3D

>>hat we=3D20

>>loop through every woman, point to her, and demand she wear an apron?

>> =20

>>

>

>It is very similar to "Every woman should wear an apron", yes.

>In Lojban you can be as precise or imprecise as you like.

>From least to most precise:

>

> ei lo ninmu cu dasni lo cragai

> =20

>

Here is where the meaning really changes. Above is a very weak claim,=20

below is much stronger. I won't assume that lo is ro, although it's not=20

ruled out. But this is normative usage, and there's a big difference=20

between a maxim the speaker feels should apply to the type, and an=20

observational statement which may or may not generalize to the type (and=20

thus lo equal ro).

> ei ro ninmu cu dasni lo cragai

> ei ro ninmu cu dasni su'o cragai

> ei ro ninmu cu dasni pa cragai

>

>And of course you can add tense:

>

> ei ro ninmu ca ca'o ca'a dasni pa cragai

>

>"Every woman should at this moment be actually wearing an apron."=20

>If you don't express the time some of them might argue that they

>wore one yesterday and so they already fulfilled their duty.

>

> =20

>

>>>>>>lo ctuca cu fendi lo selctu mu lo vo tadni

>>>>>>The teacher will divide the class

>>>>>>into five groups of four students.

>>>>>> =20

>>>>>>

>>What teacher? Any teacher? No, not any teacher. The teacher in the

>>example. Our hypothetical teacher about whom we know nothing except tha=

t

>>he's teaching Scooby Doo to a bunch of 3rd graders.=20

>> =20

>>

>

>And who might never exist, right.

> =20

>

He exists like a fictional story character. But I think we understand=20

each other here.

>>If each reference to

>>a teacher referred once again to any, non-specific teacher, each

>>sentence might refer to a bi'u teacher, rendering the script nonsensica=

l.

>>We could have given him a name, *unlike* the needed doctor.

>> =20

>>

>

>Not nonsensical, just more vague. But context will help sort it out.

>When telling a story we don't need to put a tense in every sentence,

>as usually things are told in the order they happened. You can, of

>course be more precise when you need or want to.

>

> =20

>

>>Suppose the instructions included a second teacher. Would that be any

>>teacher? Again no, because "any teacher" could include the first teache=

r.

>> =20

>>

>

>If it's important that it's about two teachers then you'd have to use

>number, of course. If it's irrelevant if there is one teacher or two

>conducting the lesson, you might not even mention number.

>=20

> =20

>

>>Furthermore, this example is totally prenexable: "Let there be a teache=

r.=3D

>> Let=3D20

>>there be a classroom..."

>> =20

>>

>

>Indeed, and there's nothing wrong with doing it that way.

>But let's get away from the idea that there is always one

>correct gadri for each situation and all the rest are wrong.

>{lo} is the most general gadri and so it will practically=20

>never be wrong in cases when another gadri is more precise.

>

>mu'o mi'e xorxes

>

>

>

>=09

> =09

>'__

>Do you Yahoo!?

>Friends. Fun. Try the all-new Yahoo! Messenger.

>http://messenger.yahoo.com/=20

>

>

> =20

>

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Tue 01 of June, 2004 23:11 GMT

1. Of course, just what THAT means is a large part of the issue here. That aside, what follows is right as I understand it (and I think I am now close to xorxes except on the issues of whether the gadri should be {lo} and whether all this has anything to do with xorxes' {lo} reported elsewhere.).

Jorge Llambías wrote:


> Jorge Llamb=EDas wrote:

> >What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

>

> I must interpret it extensionally, resulting to "some children...some=20

> adults", unchanged from the old-lo, and not a really as a maxim=20

> *because* of it's lack of ro.

Ok, but that's not the proposed lo. The proposed lo does not have

a hidden quantifier that you must glork from context. 1.All the

sentence says, with the proposed lo, is:

"It should be so that: CHILDREN forgive ADULTS"

Nothing more. That's all the statement says. How you take that to

particular instances is up to you, it is not contained in the

sentence. It says nothing about how many children should do what

to how many adults. If you want, you can add precision in different

ways. One way is to add universal quantifiers to one or both

terms. Another way is to add tense for example:

ei roroiku fe'eroroiku lo verba cu fraxu lo makcu prenu

Everytime and everywhere children should forgive adults.

> And the hypotheticality given by .ei might=20

> be enough to avoid nerdy criticism of "all". I used to think that the=20

> non-specificity of lo forced the statement to apply to the entire type,=20

> but now I don't.

The non-specificity of lo doesn't force anything about instances.

It is simply not a statement about instances.

> "Some lions hunt at night" means at least one does, and only means that=20

> we can no longer say that none do.

Of course, that's what {su'o cinfo cu kalte ca lo nicte} means.

> It doesn't tell us about the habits=20

> of lions, which I cannot see as anything but a ro statement (if part of=20

> the definition of lion) or a so'a statement (if an observed property,=20

> god forbid you should neglect the outliers).

But {lo cinfo cu kalte ca lo nicte} is meant neither as "some lions

hunt at night" nor as "all lions hunt at night". It is meant as the

more vague "lions hunt at night", without reference to the number of

instances. If you are told that, you may or may not be interested

in enquiring further, "do they hunt only at night?" "do they hunt

every night?" "does every lion hunt at night?" "does every lion

hunt every night?" "do only lions hunt at night?" and many other

questions, none of which are answered by the original claim, though

they may be suggested by the context.

The information of number of instances is just not contained in lo,

which is good because sometimes we are not interested in it. If we

make lo overprecise, we don't have a way of expresing ourselves

when precision is not possible or not wanted.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 01 of June, 2004 23:11 GMT

1. Maxims are a bad place to fight this fight sincew maxims tend to talk about aall, but also always have unmentioned exceptions, i.e., are not really about all at all. On the other hand, generic {lo} does seem to miss some of the apparent moral force of a maxim by being honest about what is covered (though possibly allowing too much).

2. Note that {ei} creates an intensional context. Within that context, however, the same rules apply as outside it, whatever they are. And, of course, maxims precisely are not derived from observation but proponded against observation (no one tells kids they ought to foregive their elders if they already do).

3. Yes, but that doesn't connect with quantifiers more than minimally; the next step — that if only one lion does it then... already is irrelevant.

Jorge Llamb=EDas wrote:

>--- xod wrote:

> =20

>

>>Jorge Llamb=3DEDas wrote:

>> =20

>>

>>>What does {ei lo verba cu fraxu lo makcu prenu} mean according to you?

>>> =20

>>>

>>I must interpret it extensionally, resulting to "some children...some=3D=

20

>>adults", unchanged from the old-lo, and not a really as a maxim=3D20

>>*because* of it's lack of ro.=20

>> =20

>>

>

>Ok, but that's not the proposed lo. The proposed lo does not have=20

>a hidden quantifier that you must glork from context. All the

>sentence says, with the proposed lo, is:

>

>"It should be so that: CHILDREN forgive ADULTS"

>

>Nothing more. That's all the statement says. How you take that to

>particular instances is up to you, it is not contained in the=20

>sentence.=20

>

1. How can it not be there from context? Part of the context of lo is that

the speaker chose to use lo instead of some other gadri. Given that

maxims are supposed to apply universally, and given that the speaker

avoided ro, and given that the speaker is being cooperative, we must

conclude that it's not a maxim. I am not saying that lo in this case is

incorrect, but that it is unhelpful.

>It says nothing about how many children should do what

>to how many adults. If you want, you can add precision in different

>ways. One way is to add universal quantifiers to one or both

>terms. Another way is to add tense for example:

>

> ei roroiku fe'eroroiku lo verba cu fraxu lo makcu prenu

> Everytime and everywhere children should forgive adults.

>

> =20

>

>>And the hypotheticality given by .ei might=3D20

>>be enough to avoid nerdy criticism of "all". I used to think that the=3D=

20

>>non-specificity of lo forced the statement to apply to the entire type,=

=3D20

>>but now I don't.

>> =20

>>

>

>The non-specificity of lo doesn't force anything about instances.

>It is simply not a statement about instances.

> =20

>

2.I have a hard time not reducing lo to an extensional claim in an

extensional context such as this. Particularly, when the statement

really should be derived from observations and falsifiable by other

observations. 3.Surely you agree that if NO lions hunt at night, then this

non-extensional claim would be false, yes?

--=20

Motorists honked in celebration in this Ramadi as news spread of the=20

assassination of the president of the Iraqi Governing Council Ezzidin Sal=

im=20

Monday. "The GC is nothing," one man shouted. "They are not the Governing=

=20

Council. They are the Prostitution Council."

Maxims are a bad plce to fight this fight



Posted by Anonymous on Tue 01 of June, 2004 23:11 GMT

1. Even generic {lo} seems to require that there are some of the things to keep from meaninglessness. But otherwise yes.

Jorge Llambías wrote:

xod:

> Jorge Llamb=EDas wrote:

> >"It should be so that: CHILDREN forgive ADULTS"

> >

> >Nothing more. That's all the statement says. How you take that to

> >particular instances is up to you, it is not contained in the=20

> >sentence.=20

>

> How can it not be there from context? Part of the context of lo is that

> the speaker chose to use lo instead of some other gadri.

1.That's not how I see it. You cannot help but to use a gadri.

Otherwise you can't make a sumti. All that {lo} does is turn

a selbri into a sumti, it adds nothing else. If you want to

add precision, you have other gadri or quantifiers for that,

but using {lo} is like using {cu}, it's vacuous. You don't

choose it, it's imposed by the grammar as the minimal

selbri-to-sumti converter.

>Given that

> maxims are supposed to apply universally, and given that the speaker

> avoided ro, and given that the speaker is being cooperative, we must

> conclude that it's not a maxim. I am not saying that lo in this case is

> incorrect, but that it is unhelpful.

The speaker did not avoid {ro}, just as when you use {cu}

instead of {ca} or {pu} you are not avoiding {ca} or {pu}.

You simply don't care to be that precise.

> >The non-specificity of lo doesn't force anything about instances.

> >It is simply not a statement about instances.

>

> I have a hard time not reducing lo to an extensional claim in an

> extensional context such as this. Particularly, when the statement

> really should be derived from observations and falsifiable by other

> observations. Surely you agree that if NO lions hunt at night, then

> this non-extensional claim would be false, yes?

Yes. If you know that {lo cinfo cu kalte ca lo nicte} you

can conclude that {su'o cinfo cu kalte ca su'o nicte}.

And you can also conclude that {lo cinfo su'oroi kalte ca

lo nicte}.

But both statements with {su'o} are more precise than the

general statement without su'o.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 01 of June, 2004 23:15 GMT

A summary of where I think we are on the first step in dealing with gadri.

Up to this point, I think that {lo} has been used in at tleast the following ways. People have also tried to express many of these positions inwith other gadri – or other devices. Gadri – indeed {lo} have been used intending to express some combinations of these as well.

Lo1 – particular (CLL)

{lo broda} refers to particular identifiable broda on a particular occasion but that identification is not specified beyond {broda}. In most respects, {lo broda} is equivalent to {su’o broda}, except that it can occur in certain environments where the other is excluded with the same meaning({lo PA broda} does not mean the same as {su’o PA broda} with the same PA. {PA lo broda} means the same as {PA broda} so long as the internal quantifier does not occur. Each occurrence of {lo broda} — whether literal or with literal pronouns, may refer to different broda – like {su’o broda}. Same-reference anaphora uses such pronouns (not yet strictly determined but that is off this point) or, typically, {le broda}.

  • Officially, the internal quantifier in {lo PA broda} is the size of the set of all broda, the external quantifier {PA (lo) broda} is the number of brodas referred to on this occasion. Officially, these default to {ro} and {su’o} respectively. Some folk have frequently said that these defaults (indeed, having defaults) were more trouble than they were worth, creating extra steps in quantifier situations and the like, and so have recommended eliminating them while elaving the option of using explicit ones as needed for meaning. Some have also suggested that reports on the size of the class of broda are generally uninformative and that other uses for internal quantifiers might be found (this is often allied with other readings of {lo}).

Lo2 – generic

{lo broda} refers to unspecified (indeed unspecifiable) broda in unspecified situations. As such, quantifiers are strictly irrelevant except that {lo broda cu brode} is entailed {ro broda cu brode} and entails {su’o broda cu brode}. No number of exceptions (short of all) to {lo broda cu brode}-- broda that don’t brode or occasions when they don’t — falsify the claim but typically single instances or a few don’t establish it either. Some cases – outstanding broda – may establish it or masses may weigh against contrary outstanding cases (truth is from weight rather than number and the propounder gets to assign the weights – not that any such assignment or even checking goes on). {lo broda cu brode} makes not claim that the behavior mentioned is common, typical, normal, average, nor weird, only that it does occur. (There is an implication that it is either common or weird just because it is mentioned, but that is not asserted – and either side of the possibility may be intended.)

In general, {PA (lo) broda} reduces to the same expression using {lo1}. Lo2 seems especially asociated with the notion that {lo PA broda} should be used as a more precise version of {lo broda PAmei}, unspecifiable groups of PA broda, considered as acting distinctly from individual broda.

In a single context, repeated occurrences of {lo2 broda} may be taken to have “the same reference” so long as not too much is made of this (like insisting that some particular broda is represented in both cases). This can lead to an occasional paradox, which must then be resolved by introducing a bit more precision, but generally one ca proceed as though the reference were constant.

Lo3 – species

{lo broda} refers to the concept of broda both in the Great Semantic Web but also in factual Weaving of the World. For example, lo ractu is semantically connected to animals, gnawers, furbearing and factually to threatening Australia and overbreeding. Quantifiers play no role here at all (well, maybe PA broda hook up differently from broda tout court and so internal quantifiers might have a role). Repeated occurrences refer to the same thing, of course. The interesting question is what kinds of sentences can this expression enter into. We want to say two kinds of things about species – species kinds of things and specimen kinds. The first is that a species is a species and that it falls under such and such genera (in the broad sense, as is “species”). The second – sometimes related – is that “members” of the species have such and such properties more or less by virtue of their membership. The second involves ordinary predicates, the first involves relations like “falls

under,” “is a subclass of,” “intersects with,” and the like. All of the second kind of things can be said using the first line of chat. “Rabbits are animals” and “The species Rabbit falls under the genus Animal” say much the same thing. But the first is much longer and the basic items here are much less commonly said. The the norm sems to be to use the second kind of locution, {lo ractu cu danlu} rather than {lo ractu cu klesi lo danlu} (or something along that line). When we need to talk species talk, it turns out to be as useful to talk set talk: {lo’i ractu cu klesi lo’i danlu}. This allows another advantage: we can circumvent the problem of saying of something that does not exist that it does not exist: the species always does exist so all we need to is say that it does not fall under lo zasti.

Since external quantifiers are irrelevant (there is exactly one of each species) , we can use them to refer to specimens {su’o lo ractu} behaves just like {lo ractu} with {lo1} (or {lo2 for that matter).

Lo4 – goo

{lo broda} refers to the substance of which individual broda are made. It akes fractional quantifiers for gobs of goo and perhaps (quasi Chinese) regular quantifiers for “natural” gobs. This is the least explored alternative and so there is less to say bout it. The goo is the same throughout but different gobs (explicit external quantifiers) may be different. Internal quantifiers do not seem to make much sense either way.

Obviously 2 and 3 with one another and parts of both with 1 have a lot in common. Many suggested usages seem to have started with that core and built on it toward one or the other of them but often incorporated inconsistent bits of the other. It may be that one or the other of 2 and 3 is enough (1 seems pretty clearly to be subsumable under either). It may even be that there is a form of 2 and 3 that are so indistinguishable tht we need not declare which we are using. Or we may need both and even 1 as well. 4 seems to be a separate case, not readily encompassing any other except perhaps 1.

Notice in passing that all of these are presented as observable real world notions, not – in any troublesome way “intensional.”



Posted by Anonymous on Tue 01 of June, 2004 23:15 GMT

Some corrections.

lo2 : {lo broda na brode} neither implies nor is implied by {lo broda naku brode}; the first fails when there are no broda, the second because of the vagueness of how many broda are required for {lo broda}. {lo broda} is also not transparent to other connectives — the contradiction problem.

lo3: Whether this is transparent depends upon just what {lo broda cu brode} means: subsumption or overlap. Each has advantages. If both {lo2} and {lo3} are used, {lo3} would be subsumption, since {lo2} (and {lo1}, which is derivable from either) covers intersections better. In that case, {lo3} is transparent to all and, indeed, behaves like a constant (which it is, after all).

To the proposal:

I do not now think that median and mode need their own gadri, since they are real things and so can be handled using {le} and some suitable predicates, probably {midju} or a lujvo on it for "median" and maybe {fadni} or a compound for "mode."



Posted by Anonymous on Tue 01 of June, 2004 23:15 GMT

pc:

> To the proposal:

> I do not now think that median and mode need their own gadri, since they are

> real things and so can be handled using {le} and some suitable predicates,

> probably {midju} or a lujvo on it for "median" and maybe {fadni} or a

> compound for "mode."

{midju} works for naturally ordered sets:

li re cu midju li pa ce li re ce li ci

2 is in the middle (is the median) of set {1,2,3}.

But for unordered sets it is less clear:

la djan cu midju lo'i prenu

John is in the middle of the set of persons.

Without an ordering for the set of persons, the above does not

seem to make sense. I guess {midju} centrally refers to spatial

position, but even then for people distributed over the surface

of the globe that is not much help.

We could use {midju} with an added {sepo'i} term:

la djan cu midju lo'i prenu sepo'i lo ka ce'u nanca makau

John is the median of the set of persons when ordered by

how old they are.

The mode makes sense for sets that can have repeated members.

The mode of {1,3,4,4,4} is 4. Maybe we could use {rapraicmi},

"x1 is the most repeated member of x2" for this:

li vo cu rapraicmi li pa ce li ci ce li vo ce li vo ce li vo

4 is the mode (most repeated member) of {1,3,4,4,4}.

{fadni} would be closer to something like: "x1 is a member of

x3 whose value by x2 is the mode", but I think it has to be a

very significant mode for fadni to work. For example, Chinese

people are the ones who have the modal value of nationality

among humans, but I'm not sure I would want to say that

all Chinese and only the Chinese are fadni in nationality.

Perhaps a relativised {fadrai}, "most typical", would work.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Re: BPFK Section: gadri


Posted by PierreAbbat on Wed 02 of June, 2004 04:29 GMT posts: 324

MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

I think {lo} should remain as it is, with these possible changes:

1. Currently {ze lo ze bidju} means "seven of the only seven beads that exist". This meaning of the use of the inner quantifier is rarely needed, so it should be dropped. If you want to say that there are only seven beads in the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group of seven beads", not "seven groups of seven beads each", which requires {zemei}. {mu lo vo tadni} is nonsense.

2. With some predicates, {lo broda} has to refer to something which may not exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't removed from her, it never formed. So {lo zunle ke jarco degji jgalu be kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its x2 may be {da'i}; if you need to say that the thing lacked does in fact exist, say {da'inai}.



Posted by Anonymous on Wed 02 of June, 2004 09:44 GMT

On Tue, Jun 01, 2004 at 09:29:13PM -0700, [email protected] wrote:

> Re: BPFK Section: gadri

> MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

>

> I think {lo} should remain as it is, with these possible changes:

> 1. Currently {ze lo ze bidju} means "seven of the only seven beads that exist". This meaning of the use of the inner quantifier is rarely needed, so it should be dropped. If you want to say that there are only seven beads in the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group of seven beads", not "seven groups of seven beads each", which requires {zemei}. {mu lo vo tadni} is nonsense.

> 2. With some predicates, {lo broda} has to refer to something which may not exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't removed from her, it never formed. So {lo zunle ke jarco degji jgalu be kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its x2 may be {da'i}; if you need to say that the thing lacked does in fact exist, say {da'inai}.

If you look at what the actual definition of {lo} is in XS, and not the

weird-ass metaphysical discussions going on about it, you'll find that it's

very reasonable: {lo} converts a selbri to a sumti without implying anything

else.

Your proposal for quantifiers is not very well-formed. The {ro ze} in {lo ro ze

bidju} is a single number, though I have no idea what the heck number it is.

I really think that this page needs to be reformulated to clarify just how

simple the new {lo} is. Right now, it doesn't draw much attention to its

simplicity, so people who have tried to follow the discussion assume that the

evil Mr. Rabbit is lurking behind the scenes.

--

Rob Speer



Posted by Anonymous on Wed 02 of June, 2004 14:47 GMT

pier:

> MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

Don't worry about it. The definitions on the page should stand on

their own.

> I think {lo} should remain as it is, with these possible changes:

> 1. Currently {ze lo ze bidju} means "seven of the only seven beads that

> exist". This meaning of the use of the inner quantifier is rarely needed, so

> it should be dropped. If you want to say that there are only seven beads in

> the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group

> of seven beads", not "seven groups of seven beads each", which requires

> {zemei}. {mu lo vo tadni} is nonsense.

It seems to me that the use of inner quantifiers you propose is

only slightly less rarely needed than the current one. Perhaps you

could give some examples (preferably natural sounding) of how

it would be used.

> 2. With some predicates, {lo broda} has to refer to something which may not

> exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji

> jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't

> removed from her, it never formed. So {lo zunle ke jarco degji jgalu be

> kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its

> x2 may be {da'i}; if you need to say that the thing lacked does in fact

> exist, say {da'inai}.

I have no problem with {la katr,in. kartrait.djonz. me'andi skagau

le degji jipno .imu'ibo claxu lo jgalu}. As far as I'm concerned

that's perfectly fine with the proposed {lo}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 14:47 GMT

Rob Speer:

> Your proposal for quantifiers is not very well-formed. The {ro ze} in {lo ro

> ze

> bidju} is a single number, though I have no idea what the heck number it is.

{roze} for "all seven" is CLL-sanctioned:

-----------------------

Another possibility is that of combining definite and indefinite

numbers into a single number. This usage implies that the two kinds

of numbers have the same value in the given context:

8.18) mi viska le rore gerku

I saw the all-of/two dogs.

I saw both dogs.

8.19) mi speni so'ici prenu

I am-married-to many/three persons.

I am married to three persons

(which is ``many in the circumstances).

Example 8.19 assumes a mostly monogamous culture by stating that

three is ``many.

----------------------

> I really think that this page needs to be reformulated to clarify just how

> simple the new {lo} is. Right now, it doesn't draw much attention to its

> simplicity, so people who have tried to follow the discussion assume that the

> evil Mr. Rabbit is lurking behind the scenes.

How do you suggest I word the definition? I think the examples

show the simplicity of the new {lo}. Just try saying them with

the old one.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 14:47 GMT

A> Fib\ne, but I think that the discussion (current and past) demonstrates the need for at least {lo2} and {lo3} somehow. Suggestions?

B> This makes more sense than the present system — and fits in with {le} at least.

C> I don't quite understand this use of {da'i}, which seems to have sentential scope, not sumti. Poc\ssibly {claxu2} creates an opque context, i.e., is some sort of abstrct description. But that does not seem to fit with the definition.

[email protected] wrote:

Re: BPFK Section: gadri

MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

A>I think {lo} should remain as it is, with these possible changes:

B>1. Currently {ze lo ze bidju} means "seven of the only seven beads that exist". This meaning of the use of the inner quantifier is rarely needed, so it should be dropped. If you want to say that there are only seven beads in the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group of seven beads", not "seven groups of seven beads each", which requires {zemei}. {mu lo vo tadni} is nonsense.

C>2. With some predicates, {lo broda} has to refer to something which may not exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't removed from her, it never formed. So {lo zunle ke jarco degji jgalu be kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its x2 may be {da'i}; if you need to say that the thing lacked does in fact exist, say {da'inai}.



Posted by Anonymous on Wed 02 of June, 2004 14:47 GMT

A> But the discussion around the paper clearly shows that Mr. Rabbit is lurking. I agree that virtually the same results can be obtained without at leaast that bit of metaphysical argle-bargle. I do think that laying the real story out would be useful somewhere.

B> Since {roze} is well-formed but with an unassigned meaning, the proposed one seems quite plausible, even if clarly related to the corresonding English.

C> Could you lay out the semantics of this simple notion.

Rob Speer wrote:

On Tue, Jun 01, 2004 at 09:29:13PM -0700, [email protected] wrote:

> Re: BPFK Section: gadri

> MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

>

> I think {lo} should remain as it is, with these possible changes:

> 1. Currently {ze lo ze bidju} means "seven of the only seven beads that exist". This meaning of the use of the inner quantifier is rarely needed, so it should be dropped. If you want to say that there are only seven beads in the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group of seven beads", not "seven groups of seven beads each", which requires {zemei}. {mu lo vo tadni} is nonsense.

> 2. With some predicates, {lo broda} has to refer to something which may not exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't removed from her, it never formed. So {lo zunle ke jarco degji jgalu be kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its x2 may be {da'i}; if you need to say that the thing lacked does in fact exist, say {da'inai}.

A>If you look at what the actual definition of {lo} is in XS, and not the

weird-ass metaphysical discussions going on about it, you'll find that it's

very reasonable: {lo} converts a selbri to a sumti without implying anything

else.

B>Your proposal for quantifiers is not very well-formed. The {ro ze} in {lo ro ze

bidju} is a single number, though I have no idea what the heck number it is.

C>I really think that this page needs to be reformulated to clarify just how

simple the new {lo} is. Right now, it doesn't draw much attention to its

simplicity, so people who have tried to follow the discussion assume that the

evil Mr. Rabbit is lurking behind the scenes.

--

Rob Speer



Posted by Anonymous on Wed 02 of June, 2004 14:48 GMT

A> I confess that I have trouble in casual reading to remember what exactly is the difference between a group of seven broda and a heptad of broda. It is the external quantifier that makes the difference, whether it is partitive or repetitive: is {ci lo ze broda} three out of the one group of seven broda or three broda heptads. I am also not sure which is the more useful. Are there stats on this? But it is clear that we can get broda heptads with the present system (or this minor modification); how do we get partititves from the heptad system(I am sure there is a straightforward way of doing it, I just don't see it off hand).

B> This remarks makes it seem that the proposed {lo} is {lo3}, whereas others more or less force {lo2}. Maybe the notion is not quite as simple as rabspir thinks.

Jorge Llambías wrote:

pier:

> MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

Don't worry about it. The definitions on the page should stand on

their own.

> I think {lo} should remain as it is, with these possible changes:

> 1. Currently {ze lo ze bidju} means "seven of the only seven beads that

> exist". This meaning of the use of the inner quantifier is rarely needed, so

> it should be dropped. If you want to say that there are only seven beads in

> the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group

> of seven beads", not "seven groups of seven beads each", which requires

> {zemei}. {mu lo vo tadni} is nonsense.

A>It seems to me that the use of inner quantifiers you propose is

only slightly less rarely needed than the current one. Perhaps you

could give some examples (preferably natural sounding) of how

it would be used.

> 2. With some predicates, {lo broda} has to refer to something which may not

> exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji

> jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't

> removed from her, it never formed. So {lo zunle ke jarco degji jgalu be

> kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its

> x2 may be {da'i}; if you need to say that the thing lacked does in fact

> exist, say {da'inai}.

B>I have no problem with {la katr,in. kartrait.djonz. me'andi skagau

le degji jipno .imu'ibo claxu lo jgalu}. As far as I'm concerned

that's perfectly fine with the proposed {lo}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 14:48 GMT

A> Nice to see CLL doing something right in anticipation of needs.

B> The examples present a number of problems, as I have noted. Most of them derive from some uncertainty about the meaning of your {lo} in that, with some one meaning each could be resolved, but that the resolving meaning appears to be different for different cases. So fixing on one relatively simple and metaphysically unsuspect meaning and then checking that all the examples fit it would be a big help. Mr. Rabbit, as presented, won't do.

Jorge Llambías wrote:

Rob Speer:

> Your proposal for quantifiers is not very well-formed. The {ro ze} in {lo ro

> ze

> bidju} is a single number, though I have no idea what the heck number it is.

A>{roze} for "all seven" is CLL-sanctioned:

-----------------------

Another possibility is that of combining definite and indefinite

numbers into a single number. This usage implies that the two kinds

of numbers have the same value in the given context:

8.18) mi viska le rore gerku

I saw the all-of/two dogs.

I saw both dogs.

8.19) mi speni so'ici prenu

I am-married-to many/three persons.

I am married to three persons

(which is ``many in the circumstances).

Example 8.19 assumes a mostly monogamous culture by stating that

three is ``many.

----------------------

> I really think that this page needs to be reformulated to clarify just how

> simple the new {lo} is. Right now, it doesn't draw much attention to its

> simplicity, so people who have tried to follow the discussion assume that the

> evil Mr. Rabbit is lurking behind the scenes.

B>How do you suggest I word the definition? I think the examples

show the simplicity of the new {lo}. Just try saying them with

the old one.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 14:48 GMT

A> Use of the statistical notions presuppose that there are statistics to cover the case (at least informal one), so we can assume whatever is needed to make this meaningful. In this case, some sort of ordering. The {sepo'i} is a good idea — or something to indicate the property involved (with an implicit ordering, e.g, salary by amount, height by height, and so on).

B> I agree that mode is trickier, since — for one thing — there can be more than one mode in a set. So the idea of a predicate only is useful ({le} then could select one for consideration or the lot of them, depending). But again, the fact that there are the statistics will simplify out the cases: the Chinese example is not, apparently, a mode but simply the most frequent value.

Jorge Llambías wrote:

pc:

> To the proposal:

> I do not now think that median and mode need their own gadri, since they are

> real things and so can be handled using {le} and some suitable predicates,

> probably {midju} or a lujvo on it for "median" and maybe {fadni} or a

> compound for "mode."

A>{midju} works for naturally ordered sets:

li re cu midju li pa ce li re ce li ci

2 is in the middle (is the median) of set {1,2,3}.

But for unordered sets it is less clear:

la djan cu midju lo'i prenu

John is in the middle of the set of persons.

Without an ordering for the set of persons, the above does not

seem to make sense. I guess {midju} centrally refers to spatial

position, but even then for people distributed over the surface

of the globe that is not much help.

We could use {midju} with an added {sepo'i} term:

la djan cu midju lo'i prenu sepo'i lo ka ce'u nanca makau

John is the median of the set of persons when ordered by

how old they are.

B>The mode makes sense for sets that can have repeated members.

The mode of {1,3,4,4,4} is 4. Maybe we could use {rapraicmi},

"x1 is the most repeated member of x2" for this:

li vo cu rapraicmi li pa ce li ci ce li vo ce li vo ce li vo

4 is the mode (most repeated member) of {1,3,4,4,4}.

{fadni} would be closer to something like: "x1 is a member of

x3 whose value by x2 is the mode", but I think it has to be a

very significant mode for fadni to work. For example, Chinese

people are the ones who have the modal value of nationality

among humans, but I'm not sure I would want to say that

all Chinese and only the Chinese are fadni in nationality.

Perhaps a relativised {fadrai}, "most typical", would work.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 14:48 GMT

pc:

> A> I confess that I have trouble in casual reading to remember what exactly

> is the difference between a group of seven broda and a heptad of broda. It

> is the external quantifier that makes the difference, whether it is partitive

> or repetitive: is {ci lo ze broda} three out of the one group of seven broda

> or three broda heptads. I am also not sure which is the more useful. Are

> there stats on this?

I don't know if there are stats. To me the obvious way to see

which is more useful is to consider the most common group

used: singletons. Quantifying over members of a singleton

is a waste of time. Quantifying over instances of a singleton

is the most common use of quantifiers.

> But it is clear that we can get broda heptads with the

> present system (or this minor modification); how do we get partititves from

> the heptad system(I am sure there is a straightforward way of doing it, I

> just don't see it off hand).

We can get both relatively easily:

PA mupli be lo ze broda

PA instances of lo ze broda

PA cmima be lo ze broda

PA members of lo ze broda

(In the case of {le} the situation is reversed. We normally have

a single instance in mind (be it of an individual or group), so

quantifying over instances is a waste of time, the useful quantification

in this case is over members when we have a specific group in mind.)

> B> This remarks makes it seem that the proposed {lo} is {lo3}, whereas

> others more or less force {lo2}. Maybe the notion is not quite as simple as

> rabspir thinks.

If lo2 does not act like a constant term, then the proposed lo

is not lo2.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 16:54 GMT

1> I doubt that the most common group is a singleton; the most common is surely no group at all but just and individual. But then I suppose that is what you meant; quantifying into a singleton would make sense, though on with {pa} — and fractionals. I again would say that the most common thing would be to count individuals, which I assume is what you mean. But it does not seem to me that that helps at all with the question of internal quantifiers as group sizes, since the analogy is not very good.

2> Yes, but these are rather complex. The first has a short form and the second does not, but I suspect the second is more common — or at least as common — as the first.

3> {lo2} does not behave like a constant, since it is not one (it is not tranparent to any operation). {lo3} does behave like a constant (since it is one) but is abstract and relatively impervious (as described so far) to factual properties (though that could be changed fairly easily at this point).

Jorge Llambías wrote:

pc:

> A> I confess that I have trouble in casual reading to remember what exactly

> is the difference between a group of seven broda and a heptad of broda. It

> is the external quantifier that makes the difference, whether it is partitive

> or repetitive: is {ci lo ze broda} three out of the one group of seven broda

> or three broda heptads. I am also not sure which is the more useful. Are

> there stats on this?

1>I don't know if there are stats. To me the obvious way to see

which is more useful is to consider the most common group

used: singletons. Quantifying over members of a singleton

is a waste of time. Quantifying over instances of a singleton

is the most common use of quantifiers.

> But it is clear that we can get broda heptads with the

> present system (or this minor modification); how do we get partititves from

> the heptad system(I am sure there is a straightforward way of doing it, I

> just don't see it off hand).

2>We can get both relatively easily:

PA mupli be lo ze broda

PA instances of lo ze broda

PA cmima be lo ze broda

PA members of lo ze broda

(In the case of {le} the situation is reversed. We normally have

a single instance in mind (be it of an individual or group), so

quantifying over instances is a waste of time, the useful quantification

in this case is over members when we have a specific group in mind.)

> B> This remarks makes it seem that the proposed {lo} is {lo3}, whereas

> others more or less force {lo2}. Maybe the notion is not quite as simple as

> rabspir thinks.

3>If lo2 does not act like a constant term, then the proposed lo

is not lo2.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 16:54 GMT

pc:

> 1> I doubt that the most common group is a singleton; the most common is

> surely no group at all but just and individual. But then I suppose that is

> what you meant;

Yes, I meant {lo pa broda}, with the proposed {lo}.

> quantifying into a singleton would make sense, though on with

> {pa} — and fractionals.

That would be:

PA pagbu be lo pa broda

That's a possible use for fractional quantifiers, though not

my preferred one.

> I again would say that the most common thing would

> be to count individuals, which I assume is what you mean. But it does not

> seem to me that that helps at all with the question of internal quantifiers

> as group sizes, since the analogy is not very good.

I understand Pierre does not object to using internal quantifiers

as group sizes. He objects to the use of the external quantifiers to

quantify over instances rather than over members.

I guess {PA broda} can equally well be understood as quantifying

over members of the group of all broda, or over instances of

a single broda.

(Re:mupli & cmima)

> 2> Yes, but these are rather complex. The first has a short form and the

> second does not, but I suspect the second is more common — or at least as

> common — as the first.

I presented some examples with instances of groups. Perhaps if

someone presented some examples of members of (generic) groups

we could get a better idea of what we are comparing.

> 3> {lo2} does not behave like a constant, since it is not one (it is not

> tranparent to any operation). {lo3} does behave like a constant (since it is

> one) but is abstract and relatively impervious (as described so far) to

> factual properties (though that could be changed fairly easily at this

> point).

If lo3 could be made more pervious to factual properties, then

we may be converging.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Re: BPFK Section: gadri


rlpowell Posted by rlpowell on Wed 02 of June, 2004 16:55 GMT posts: 14214

Test post; Pierre is apparently having problems.

As a side note, this thread in the forums takes a long time to load. This is primarily because it has reached almost 1MiB in size.

-Robin



Posted by Anonymous on Wed 02 of June, 2004 17:39 GMT

A> This doesn't help me much, since I have understood {lo pa broda} to be, like the other {lo PA broda}, about a group with PA members (as your expansion suggests). {PA1 lo PA2 broda} is then PA1 distinct (not necessary separate) groups with PA2 members, eqquivalent to your {PA1 mupli be lo PA2 broda} (stretching {mupli} somewhat perhaps). To refer to, say, two of these guys requires something like {refe'iPA2 lo PA2 broda}, or — if their being in this group is not important — just {re broda} (but not, obviously, {re lo broda}), equivalent to your {re cmima be be lo PA2 broda}. So, to talk about a single individual, one has to say {pa broda}. You suggest that pierre is OK with your use of internal quantifers but wants the external to be used in the {cmima} sense. Presumably your external quantifer sense ({mupli}) would, for him be covered by {PA1lo broda PA2mei} — or maybe without the {lo}. Sounds like a Zipfean question; any ideas how to sort matters out?

B> I am not clear what part of a set is, even a singleton, so I suppose this is part of an individual, {PA broda} with fractional PA (a controversial point in its own right, if I remember rightly).

C> Ambigous: do you mean "takes as value individuals from the set of broda" or "for some individual broda, takes as values instances of that individual" I don't quite know what you might mean by an instance of an individual, so I suspect you mean the former. But that is not different from the first quantifier reading, so I don't understand the choice you are offering.

D> Again, I don't understand what the choice here is. What is a generic group (what more so that the/a group of broda)?

E> That is, I take it, that I am getting close to understanding what you have in mind. But I had just conmvinced myself that you werre after {lo2} and had made one small mistake. Now apparently you are after {lo3} with a few complexities. If {lo3} is made pervious to claims, then every sentence {lo broda cu brode} is three-way ambiguous (with a bunch of cases where the ambiguity disappers immediately and most others easily resolved by context): 1) the species broda falls under the genus brode, 2) the species broda overlaps the genus brode, or 3) specimens of broda do brode. Forms with explicit tense fall into the last category, of course. Contradictory claims ({lo broda cu ga brode ginai brode} fall into the second, as do claims that are otherwise impossible (though some of these may be of type 1 when there are no broda: {lo pavyseljirna cu xanri danlu}. {lo broda cu brode} does not generalize to {da brode}, since, in this line of talk, species are not things (we can shift

into species talk but it is much messier), nor even, in types 1 and 2, to {su'o broda cu brode}. I suspect that this is crucially different from your {lo} and so am no further along than I was.

Jorge Llambías wrote:

pc:

> 1> I doubt that the most common group is a singleton; the most common is

> surely no group at all but just an individual. But then I suppose that is

> what you meant;

A>Yes, I meant {lo pa broda}, with the proposed {lo}.

> quantifying into a singleton would make sense, though on with

> {pa} — and fractionals.

That would be:

B>PA pagbu be lo pa broda

That's a possible use for fractional quantifiers, though not

my preferred one.

> I again would say that the most common thing would

> be to count individuals, which I assume is what you mean. But it does not

> seem to me that that helps at all with the question of internal quantifiers

> as group sizes, since the analogy is not very good.

I understand Pierre does not object to using internal quantifiers

as group sizes. He objects to the use of the external quantifiers to

quantify over instances rather than over members.

C>I guess {PA broda} can equally well be understood as quantifying

over members of the group of all broda, or over instances of

a single broda.

(Re:mupli & cmima)

> 2> Yes, but these are rather complex. The first has a short form and the

> second does not, but I suspect the second is more common — or at least as

> common — as the first.

D>I presented some examples with instances of groups. Perhaps if

someone presented some examples of members of (generic) groups

we could get a better idea of what we are comparing.

> 3> {lo2} does not behave like a constant, since it is not one (it is not

> tranparent to any operation). {lo3} does behave like a constant (since it is

> one) but is abstract and relatively impervious (as described so far) to

> factual properties (though that could be changed fairly easily at this

> point).

E>If lo3 could be made more pervious to factual properties, then

we may be converging.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 17:55 GMT

On Wed, Jun 02, 2004 at 05:43:14AM -0700, Jorge Llamb?as wrote:

> Rob Speer:

> > I really think that this page needs to be reformulated to clarify

> > just how simple the new {lo} is. Right now, it doesn't draw much

> > attention to its simplicity, so people who have tried to follow the

> > discussion assume that the evil Mr. Rabbit is lurking behind the

> > scenes.

>

> How do you suggest I word the definition? I think the examples show

> the simplicity of the new {lo}. Just try saying them with the old one.

If Rob is correct about the total lack of other implication, and/or that

the new lo is completely generic (which I think is the same thing), it

would be a good idea to explicitely point this out.

BTW, I don't think you've talked about tu'o on the page yet.

-Robin


Re: BPFK Section: gadri


Posted by PierreAbbat on Wed 02 of June, 2004 18:06 GMT posts: 324

MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and the like.

I think {lo} should remain as it is, with these possible changes:

1. Currently {ze lo ze bidju} means "seven of the only seven beads that exist". This meaning of the use of the inner quantifier is rarely needed, so it should be dropped. If you want to say that there are only seven beads in the world, say {lo ro ze bidju}. {ze lo ze bidju} then means "all of a group of seven beads", not "seven groups of seven beads each", which requires {zemei}. {mu lo vo tadni} is nonsense.

2. With some predicates, {lo broda} has to refer to something which may not exist. For instance, {la katr,in. kartrait.djonz. me'andi skagau le degji jipno .imu'ibo claxu lo jgalu}. She's missing the left index nail; it wasn't removed from her, it never formed. So {lo zunle ke jarco degji jgalu be kykydy} has no referent, and yet she claxu it. Thus {claxu} implies that its x2 may be {da'i}; if you need to say that the thing lacked does in fact exist, say {da'inai}.



Posted by Anonymous on Wed 02 of June, 2004 18:21 GMT

pier:

> Re: BPFK Section: gadri

> MEGO! I am lost trying to make sense of intensions, Mr. Rabbit, and

> the like.

Then don't read that crap. PC's out of his mind; I'm certainly not

going to try to follow it. Just work on the definitions as stated.

> I think {lo} should remain as it is, with these possible changes:

Neither of your changes allow me to say "Cats eat mice" or "I need a

doctor".

-Robin



Posted by Anonymous on Wed 02 of June, 2004 18:55 GMT

pc:

> Again, I don't understand what the choice here is. What is a generic

> group (what more so that the/a group of broda)?

For example, for {ci lo mu broda}, the choice is between "three groups

of five broda" and "three members of any group of five broda".

I want some example of use for the second case, to judge how useful

it might be. I already gave some examples of the first case.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 18:55 GMT

On Wed, Jun 02, 2004 at 11:30:44AM -0700, Jorge Llamb?as wrote:

> I want some example of use for the second case, to judge how useful it

> might be. I already gave some examples of the first case.

ci lo mu mikce cu zmanei la snaikoil

-Robin



Posted by Anonymous on Wed 02 of June, 2004 19:22 GMT

Robin Lee Powell:

> If Rob is correct about the total lack of other implication, and/or that

> the new lo is completely generic (which I think is the same thing), it

> would be a good idea to explicitely point this out.

How about:

"Generic article. The resulting expression refers generically

to any individual or group that satisfies the predicate."

> BTW, I don't think you've talked about tu'o on the page yet.

I have, in one of the notes. I didn't want to include {tu'o}

in the definition of {lo} because it is rather marginal. I doubt

it will be used much if at all. Substances don't really

need special marking in most cases. Since this is really

about {tu'o} more than about {lo}, it should be treated in

the definition of {tu'o}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 19:59 GMT

Robin Lee Powell:

> ci lo mu mikce cu zmanei la snaikoil

If we adopt And's proposal for the interpretation of fractional

quantifiers (which I think we should) then that would be:

ci fi'u mu mikce cu zmanei la snaikoil

or equivalently:

xano ce'i mikce cu zmanei la snaikoil

But I'm not sure if Pierre would read it as three out of

every five doictors, or three from some unidentified group

of five doctors.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 19:59 GMT

How do you say these sentences in XS-Lojban?

- People are alone, but there are many of them.

- People kill each other.

--=20

Arnt Richard Johansen http://arj.nvg.org=

/

Tusener p=E5 tusener av nydelige l=F8penoter - og hvilenoter!



Posted by Anonymous on Wed 02 of June, 2004 19:59 GMT


> How do you say these sentences in XS-Lojban?

>

> - People are alone, but there are many of them.

lo prenu cu nonkansa gi'eku'i so'imei

> - People kill each other.

lo prenu cu catra py

People kill people.

If you mean "each other" literally, then:

lo prenu cu simxu lo ka ce'u ce'u catra

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 20:20 GMT

On Wed, 2 Jun 2004, Jorge Llamb=EDas wrote:

> --- Arnt Richard Johansen wrote:

> > How do you say these sentences in XS-Lojban?

> >

> > - People are alone, but there are many of them.

>

> lo prenu cu nonkansa gi'eku'i so'imei

That was _not_ what I was trying to make you say! :-)

Here, try this instead:

There are many single bears in the desert, but whole packs of them in t=

he

forest.

> > - People kill each other.

>

> lo prenu cu catra py

> People kill people.

>

> If you mean "each other" literally, then:

>

> lo prenu cu simxu lo ka ce'u ce'u catra

Okay. That's actually rather nice.

--=20

Arnt Richard Johansen http://arj.nvg.org=

/

Twinkle, twinkle little star

I don't wonder what you are,

=46or by spectroscopic ken,

I know that you are hydrogen.

--Ian Bush



Posted by Anonymous on Wed 02 of June, 2004 20:20 GMT

On Wednesday 02 June 2004 15:33, Jorge "Llamb=EDas" wrote:

> If we adopt And's proposal for the interpretation of fractional

> quantifiers (which I think we should) then that would be:

>

> ci fi'u mu mikce cu zmanei la snaikoil

>

> or equivalently:

>

> xano ce'i mikce cu zmanei la snaikoil

>

> But I'm not sure if Pierre would read it as three out of

> every five doictors, or three from some unidentified group

> of five doctors.

I read it as 3/5 of a mass of doctors, which properly speaking is {ci fi'u =

mu=20

loi mikce}. But it could also mean "3/5 of a doctor" {ci fi'u mu lo mikce},=

=20

but that is nonsense in this context. {ci lo mu mikce} means three of a gro=

up=20

of five doctors.

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 02 of June, 2004 20:29 GMT

Arnt Richard Johansen:

> There are many single bears in the desert, but whole

> packs of them in the forest.

Without anaphora:

so'i lo pa bersa cu xabju lo kutytu'a

iku'i so'o lo so'i bersa cu go'i lo ricfoi

With anaphora I would say:

so'i lo pa bersa cu xabju lo kutytu'a

iku'i so'o lo so'iboi by cu go'i lo ricfoi

That {boi} should be made elidable.

(I took "the desert" and "the forest" as generic, not referring

to some particular desert or forrest you may have in mind.)

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 20:34 GMT

Pierre:

> {ci lo mu mikce} means three of a group

> of five doctors.

That's what I thought you meant.

In what kind of context would it be useful to

say something like that, where "a group of five doctors"

is not certain group that you have in mind?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 21:08 GMT

Ah so. I guess I was thinking of having hit upon a group of five broda and then taaking three out of that. If we keep the {lo mu broda} as "any group of five broda" rather than "a group of five broda, then it seems there is no use for the partitive sense. I suppose that the second sense ("a group") is to be {su'o lo mu broda} or some such and the partitive gotten at using {lu'a} (or something like it for the right sort of entity).

s wrote:

For example, for {ci lo mu broda}, the choice is between "three groups

of five broda" and "three members of any group of five broda".

I want some example of use for the second case, to judge how useful

it might be. I already gave some examples of the first case.



Posted by Anonymous on Wed 02 of June, 2004 21:08 GMT

A> This looks like {lo2}; I thought you were after {lo3}

B> What does {tu'o} have to do with substances?

Jorge Llambías wrote:

A>How about [[for%20a%20definition%20of%20%7Blo%7D|for a definition of {lo}]]:

"Generic article. The resulting expression refers generically

to any individual or group that satisfies the predicate."

.. I didn't want to include {tu'o}

in the definition of {lo} because it is rather marginal. I doubt

it will be used much if at all. Substances don't really

need special marking in most cases. Since this is really

about {tu'o} more than about {lo}, it should be treated in

the definition of {tu'o}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 21:08 GMT

pc:

> What does {tu'o} have to do with substances?

Substances don't have cardinality, so we use {tu'o}

as inner quantifier to emphasize that no number

applies there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 21:08 GMT

At some point I gathered (I thought you said it in fact) that {PA broda} was about instances, not generalities. So these critters look like some 3/5 of some doctor.

Jorge Llambías wrote:

Robin Lee Powell:

> ci lo mu mikce cu zmanei la snaikoil

If we adopt And's proposal for the interpretation of fractional

quantifiers (which I think we should) then that would be:

ci fi'u mu mikce cu zmanei la snaikoil

or equivalently:

xano ce'i mikce cu zmanei la snaikoil

But I'm not sure if Pierre would read it as three out of

every five doictors, or three from some unidentified group

of five doctors.



Posted by Anonymous on Wed 02 of June, 2004 21:08 GMT

OK but pointless, assuming we have a way to refer to substances at all, Is there one in your system at this time?

Jorge Llambías wrote:

pc:

> What does {tu'o} have to do with substances?

Substances don't have cardinality, so we use {tu'o}

as inner quantifier to emphasize that no number

applies there.



Posted by Anonymous on Wed 02 of June, 2004 21:17 GMT

pc:

> At some point I gathered (I thought you said it in fact) that {PA broda} was

> about instances, not generalities. So these critters look like some 3/5 of

> some doctor.

You may want to check the full proposal for fractional quantifiers,

by And. You seemed to like it at the time. It can be found here:

XS gadri proposal: And's version

Briefly: {PA broda} is equivalent to {PA fi'u ro broda} and is

about instances, "PA out of all". {PA1 fi'u PA2} is "PA1 out of

every PA2".

This is more about the nitty gritty of quantifiers than about gadri,

so I don't think it needs to be added to the gadri definitions. It

can be voted separately, when we do quantifiers.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 02 of June, 2004 21:17 GMT

pc:

> OK but pointless, assuming we have a way to refer to

> substances at all, Is there one in your system at this time?

{lo} can be used with substances or non-substances, it does

not by itself force a distinction. When you use an inner

quantifier, you indicate something countable, so not a

substance. When you use an explicit tu'o as inner quantifier,

you indicate a substance. In general such an explicit

indication is not necessary, but it is available.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 03 of June, 2004 00:00 GMT

This will be nice if it works, but several attempts to reduce {lo4} to {lo3} or {lo2} crahed and burnt. I don't remember details, but it seems plausible to me at least in the case of {lo2}, which is inherently quantificational. With {lo3} I haven't thought through (or remembered) how thespecies is related to its embodiment. It may be that {tu'o} is enough of a distinction. External quantifiers are then on globs? (I think BTW that something has to be said about quantifiers in the gadri section, though maybe not at this time).

Jorge Llambías wrote:

pc:

> OK but pointless, assuming we have a way to refer to

> substances at all, Is there one in your system at this time?

{lo} can be used with substances or non-substances, it does

not by itself force a distinction. When you use an inner

quantifier, you indicate something countable, so not a

substance. When you use an explicit tu'o as inner quantifier,

you indicate a substance. In general such an explicit

indication is not necessary, but it is available.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 03 of June, 2004 00:01 GMT

On Wed, Jun 02, 2004 at 01:28:33PM -0700, Jorge Llamb?as wrote:

> With anaphora I would say:

>

> so'i lo pa bersa cu xabju lo kutytu'a

> iku'i so'o lo so'iboi by cu go'i lo ricfoi

>

> That {boi} should be made elidable.

I will fight to the death against any attempt to make BY invalid in

numerical expressions. That may or may not be relevant here, but I

wanted you to know.

-Robin



Posted by Anonymous on Thu 03 of June, 2004 00:01 GMT

On Wed, Jun 02, 2004 at 12:01:47PM -0700, Jorge Llamb?as wrote:

>

> Robin Lee Powell:

> > If Rob is correct about the total lack of other implication, and/or

> > that the new lo is completely generic (which I think is the same

> > thing), it would be a good idea to explicitely point this out.

>

> How about:

> "Generic article. The resulting expression refers generically to any

> individual or group that satisfies the predicate."

Err. Nevermind.

> > BTW, I don't think you've talked about tu'o on the page yet.

>

> I have, in one of the notes.

Err, nevermind again.

-Robin



Posted by Anonymous on Thu 03 of June, 2004 02:13 GMT

Robin Lee Powell:

> On Wed, Jun 02, 2004 at 01:28:33PM -0700, Jorge Llamb?as wrote:

> > iku'i so'o lo so'iboi by cu go'i lo ricfoi

> >

> > That {boi} should be made elidable.

>

> I will fight to the death against any attempt to make BY invalid in

> numerical expressions. That may or may not be relevant here, but I

> wanted you to know.

{boi} is demonstrably a nuisance in such cases, whereas lerfu within

numbers have not as yet seen any use. I don't think I'd go so far as

to kill for it, however.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 03 of June, 2004 20:59 GMT

Withdrawn (or resolved) objections to the current proposal on {lo}:

1. The new {lo} will have essentially two different meanings with and

without quantifiers.

Icky and intellectual unfulfilling, but not really a problem. {ku} and

{bo} already have different meanings in different contexts.

2. Old uses of {lo} will have a new meaning, that will in many cases not

have been intended by the original users.

I stand corrected that most of the previous uses, even Red Book

examples, *do* seem to run counter to my understanding of it, and

apparently also to the prescription in the Red Book.

Jorge's argument that post-BPFK Lojban will look very different if an

entirely new article, which will probably be very high in frequency, is

another of the reasons that I am withdrawing this objections.

My objections to the current proposal that are as yet unresolved:

3. {tu'o} as an inner quantifier is, as I understand it, either a special

case that magically turns {lo} into a generic mass article, or else is

intended to be a part of the general quantifier system.

If the former is the case, I can't stand it, and it has to go.

If the latter, it does not appear to fit stringently into the system.

It appears to be one of these clever tricks that immediately make

sense, but does not really hold when people try to think inside the

system, instead of standing on the outside1.


1 Rather like JCB's use of the equivalent of "mutce krinu nanmu" for

"very reasonable man".

--

Arnt Richard Johansen http://arj.nvg.org/

"This is the sort of bloody nonsense up with which I will not put!"

--attributed to Winston Churchill



Posted by Anonymous on Thu 03 of June, 2004 21:11 GMT

Arnt Richard Johansen wrote:

>My objections to the current proposal that are as yet unresolved:

>

>3. {tu'o} as an inner quantifier is, as I understand it, either a special

> case that magically turns {lo} into a generic mass article, or else is

> intended to be a part of the general quantifier system.

>

>

It's not specific to lo; it could be used with le.

How would you interpret tu'o in the place of a quantifier? When do you

think quantification becomes meaningless?

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 03 of June, 2004 21:35 GMT

arj:

> 1. The new {lo} will have essentially two different meanings with and

> without quantifiers.

>

> Icky and intellectual unfulfilling, but not really a problem. {ku} and

> {bo} already have different meanings in different contexts.

I don't think the meaning of {lo} changes when you quantify over

instances. {PA lo broda} is essentially {PA mupli be lo broda}, or

{PA da poi mupli lo broda}: {lo broda} tells you the kind you're

talking about, and the quantifiers run over the instances. But

ignore this comment if it causes more confusion, since you are

not presenting this as an objection.

> My objections to the current proposal that are as yet unresolved:

>

> 3. {tu'o} as an inner quantifier is, as I understand it, either a special

> case that magically turns {lo} into a generic mass article, or else is

> intended to be a part of the general quantifier system.

{tu'o} is not mentioned in the definition of {lo}, not even

in the examples. Its meaning as an inner quantifier has to be

proposed and voted on when defining {tu'o}, that's why I only

mentioned it under "notes".

> If the latter, it does not appear to fit stringently into the system.

> It appears to be one of these clever tricks that immediately make

> sense, but does not really hold when people try to think inside the

> system, instead of standing on the outside1.

If you don't think it fits into the system, vote against it when

it is proposed in the definition of {tu'o}.

For {lo}, all you need to know is that {lo solji} can refer to

"gold" generically. Inner quantifiers are not obligatory, and

none is assumed by default when none is presented, just as

with outer quantifiers. Do you see that as problematic?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 00:15 GMT

As I understand it working within the system, since {tu'o} is the null quantifiers it indicates that NO Quantifier applies and this seems to be the case only with substances. Anything else justifies {pa} or som other number or at least {su'o} or {ro}. Now, of course, that does not work as well with the new interpretation of internal quantifiers, since they are about the size of selected groups not about the whole of the class (and so more like quantifiers with {le}), but it is still the case that only with substance do (whole-number, cardinal) quantifiers make no sense. I am not sure whether this line of reasoning is sufficient, but it is a cheap way to get substances without a new gadri.

xod wrote:Arnt Richard Johansen wrote:

>My objections to the current proposal that are as yet unresolved:

>

>3. {tu'o} as an inner quantifier is, as I understand it, either a special

> case that magically turns {lo} into a generic mass article, or else is

> intended to be a part of the general quantifier system.

>

>

It's not specific to lo; it could be used with le.

How would you interpret tu'o in the place of a quantifier? When do you

think quantification becomes meaningless?

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 04 of June, 2004 00:21 GMT

A> But unquantified {lo broda} is precisely not about instances (in one fairly frequent version of this tale), so the meaning DOES change when quantifiers are added. I am not sure that this is an objection; it is certainly a mild one compared with other problems with this tale (if we are still on the old Mr line).

B> Not with {solji}, a substance word to begin with, but please don't tell me that {lo bakni} is cow goo along with everything else.

Jorge Llambías wrote:

arj:

> 1. The new {lo} will have essentially two different meanings with and

> without quantifiers.

>

> Icky and intellectual unfulfilling, but not really a problem. {ku} and

> {bo} already have different meanings in different contexts.

A>I don't think the meaning of {lo} changes when you quantify over

instances. {PA lo broda} is essentially {PA mupli be lo broda}, or

{PA da poi mupli lo broda}: {lo broda} tells you the kind you're

talking about, and the quantifiers run over the instances. But

ignore this comment if it causes more confusion, since you are

not presenting this as an objection.

> My objections to the current proposal that are as yet unresolved:

>

> 3. {tu'o} as an inner quantifier is, as I understand it, either a special

> case that magically turns {lo} into a generic mass article, or else is

> intended to be a part of the general quantifier system.

{tu'o} is not mentioned in the definition of {lo}, not even

in the examples. Its meaning as an inner quantifier has to be

proposed and voted on when defining {tu'o}, that's why I only

mentioned it under "notes".

> If the latter, it does not appear to fit stringently into the system.

> It appears to be one of these clever tricks that immediately make

> sense, but does not really hold when people try to think inside the

> system, instead of standing on the outside1.

If you don't think it fits into the system, vote against it when

it is proposed in the definition of {tu'o}.

B>For {lo}, all you need to know is that {lo solji} can refer to

"gold" generically. Inner quantifiers are not obligatory, and

none is assumed by default when none is presented, just as

with outer quantifiers. Do you see that as problematic?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 00:34 GMT

pc:

> Not with {solji}, a substance word to begin with, but please don't tell me

> that {lo bakni} is cow goo along with everything else.

Only to the extent that cow goo does bakni. lo bakni can be

anything that baknis. For {bakni}, the goo is probably a marginal

sense, if allowed at all. If {lo se citka be mi cu bakni}, then

{mi citka lo bakni}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 14:58 GMT

Well, if what I eat is a (part of a) cow then this is not problematic, but I don't see it as meaning what I eat is cow. I suspect that the lack of quantifiers (i.e., bare {lo}) here covers the possibility of fractional ones as well as integral and indefinites. That would seem to be the natural understanding of indefiniteness.

Jorge Llambías wrote:

pc:

> Not with {solji}, a substance word to begin with, but please don't tell me

> that {lo bakni} is cow goo along with everything else.

Only to the extent that cow goo does bakni. lo bakni can be

anything that baknis. For {bakni}, the goo is probably a marginal

sense, if allowed at all. If {lo se citka be mi cu bakni}, then

{mi citka lo bakni}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 14:59 GMT

On Thursday 03 June 2004 20:30, Jorge "Llamb=EDas" wrote:

> Only to the extent that cow goo does bakni. lo bakni can be

> anything that baknis. For {bakni}, the goo is probably a marginal

> sense, if allowed at all. If {lo se citka be mi cu bakni}, then

> {mi citka lo bakni}.

I would say {loi bakni} for cow all over the road. For food, {mi citka lo=20

bakyrectu}; I wouldn't say {mi citka lo bakni} unless I ate the whole cow, =

or=20

at least as much as is edible.

Some words could refer to individuals or substances. {panono me'andi cu zva=

ti=20

le foldi} vs. {lo grake be li panono me'andi cu nenri le dakli}. An Indian =

is=20

more likely to think of {loi me'andi} as plants considered as a mass; an=20

American is more likely to think of a mass of powder.

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 14:59 GMT

pier:

> For food, {mi citka lo bakyrectu}; I wouldn't say {mi citka lo bakni}

> unless I ate the whole cow, or at least as much as is edible.

Me too. I would say {lo se citka be mi cu bakyrectu} and

{mi citka lo bakyrectu}.

The question of whether or not one could also just say {mi citka

lo bakni} has little to do with {lo}. It only concerns the meaning

of {bakni}. Just like the question about {lo se citka be mi cu bakni}

does not concern {cu}, it concerns {bakni}.

{lo} does not add meaning, it is a purely syntactical marker,

like {cu}.

> Some words could refer to individuals or substances. {panono me'andi cu zva=

> ti=20

> le foldi} vs. {lo grake be li panono me'andi cu nenri le dakli}. An Indian =

> is=20

> more likely to think of {loi me'andi} as plants considered as a mass; an=20

> American is more likely to think of a mass of powder.

That's more to do with the meaning of {me'andi} than about gadri.

lo me'andi is that which me'andis, whatever that is, that's all.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 17:05 GMT

Pierre Abbat wrote:

>On Thursday 03 June 2004 20:30, Jorge "Llamb=EDas" wrote:

>

>

>>Only to the extent that cow goo does bakni. lo bakni can be

>>anything that baknis. For {bakni}, the goo is probably a marginal

>>sense, if allowed at all. If {lo se citka be mi cu bakni}, then

>>{mi citka lo bakni}.

>>

>>

>

>I would say {loi bakni} for cow all over the road.

>

And it could easily be interpreted as a herd of cows standing in the

road. Therefore I suggest so'o bakni for cow herd, and tu'o bakni for

cow goo, and the disuse of loi.

>For food, {mi citka lo=20

>bakyrectu}; I wouldn't say {mi citka lo bakni} unless I ate the whole cow, =

>or=20

>at least as much as is edible.

>

>

lo bakni can refer to part of a cow.

Context should *always* imply that you didn't consume the entire cow,

and you shouldn't need to emphasize that except in the most bizarre

circumstances.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 04 of June, 2004 17:05 GMT

A>Well, it is about {cu}, too: am I now eating a cow, could I eat a cow, did I get through a cow over a period of time, and so on, just like the issue of whether it is one cow or several or bits and pieces of perhaps several cows.

B> At least says that he thing(s) exist(s) in the implied world. Of course, {cu} says that the event happens in the implied world, so they are on a par.

Jorge Llambías wrote:

pier:

> For food, {mi citka lo bakyrectu}; I wouldn't say {mi citka lo bakni}

> unless I ate the whole cow, or at least as much as is edible.

Me too. I would say {lo se citka be mi cu bakyrectu} and

{mi citka lo bakyrectu}.

A>The question of whether or not one could also just say {mi citka

lo bakni} has little to do with {lo}. It only concerns the meaning

of {bakni}. Just like the question about {lo se citka be mi cu bakni}

does not concern {cu}, it concerns {bakni}.

B>{lo} does not add meaning, it is a purely syntactical marker,

like {cu}.

> Some words could refer to individuals or substances. {panono me'andi cu zva=

> ti=20

> le foldi} vs. {lo grake be li panono me'andi cu nenri le dakli}. An Indian =

> is=20

> more likely to think of {loi me'andi} as plants considered as a mass; an=20

> American is more likely to think of a mass of powder.

That's more to do with the meaning of {me'andi} than about gadri.

lo me'andi is that which me'andis, whatever that is, that's all.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 17:05 GMT

  • {me'andi}? Apparently not a lujvo.

Pierre Abbat wrote:On Thursday 03 June 2004 20:30, Jorge "Llamb=EDas" wrote: > Only to the extent that cow goo does bakni. lo bakni can be

> anything that baknis. For {bakni}, the goo is probably a marginal

> sense, if allowed at all. If {lo se citka be mi cu bakni}, then

> {mi citka lo bakni}.

I would say {loi bakni} for cow all over the road. For food, {mi citka lo=20

bakyrectu}; I wouldn't say {mi citka lo bakni} unless I ate the whole cow, =

or=20

at least as much as is edible.

Some words could refer to individuals or substances. {panono me'andi cu zva=

ti=20

le foldi} vs. {lo grake be li panono me'andi cu nenri le dakli}. An Indian =

is=20

more likely to think of {loi ***me'andi} as plants considered as a mass; an=20

American is more likely to think of a mass of powder.

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 17:53 GMT

On Friday 04 June 2004 12:49, John E Clifford wrote:

> *** {me'andi}? Apparently not a lujvo.

Fu'ivla from Indic languages meaning "henna". See http://www.mehandi.com.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 17:53 GMT

On Friday 04 June 2004 11:35, xod wrote:

> And it could easily be interpreted as a herd of cows standing in the

> road. Therefore I suggest so'o bakni for cow herd, and tu'o bakni for

> cow goo, and the disuse of loi.

{so'o bakni} is equivalent to {so'o lo bakni} which refers to several cows

considered individually. {loi bakni}, referring to the same cows, considers

them as a mass.

> lo bakni can refer to part of a cow.

I don't see that in the definition of {bakni}.

> Context should *always* imply that you didn't consume the entire cow,

> and you shouldn't need to emphasize that except in the most bizarre

> circumstances.

{mi} can have a plural referent considered as a mass, in which case {mi citka

lo bakni} can be true, with all of us together eating the cow.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 18:17 GMT

Pierre Abbat wrote:

>On Friday 04 June 2004 11:35, xod wrote:

>

>

>>And it could easily be interpreted as a herd of cows standing in the

>>road. Therefore I suggest so'o bakni for cow herd, and tu'o bakni for

>>cow goo, and the disuse of loi.

>>

>>

>

>{so'o bakni} is equivalent to {so'o lo bakni} which refers to several cows

>considered individually. {loi bakni}, referring to the same cows, considers

>them as a mass.

>

>

If loi is never used for cow herds or cow goo, but only in cases where

the cows are acting collectively (doing something as a group that they

could not do individually) then loi has a useful and clear meaning. I

can't imagine an example of cows working cooperatively and not as a

number of individuals. The classic case of this is the 3 men carrying

the piano; a collective task because no one individual can do it, thus

the ability to carry a piano is an emergent property of the collective.

>>lo bakni can refer to part of a cow.

>>

>>

>

>I don't see that in the definition of {bakni}.

>

>

Half a cow is {.5 bakni}. Two cows is {2 bakni}. If {lo bakni} can refer

to the latter, then why not the former?

>>Context should *always* imply that you didn't consume the entire cow,

>>and you shouldn't need to emphasize that except in the most bizarre

>>circumstances.

>>

>>

>

>{mi} can have a plural referent considered as a mass, in which case {mi citka

>lo bakni} can be true, with all of us together eating the cow.

>

>

Yes, when mi is referring to a group with the ability to consume a cow,

then you should be clear to distinguish whether they did devour one

animal, or simply sit down to a steak meal.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 04 of June, 2004 18:24 GMT

Pierre Abbat:

> On Friday 04 June 2004 11:35, xod wrote:

> > lo bakni can refer to part of a cow.

>

> I don't see that in the definition of {bakni}.

We all agree that {lo bakni} is only for things that do bakni.

That's all that matters for the gadri discussion.

> > Context should *always* imply that you didn't consume the entire cow,

> > and you shouldn't need to emphasize that except in the most bizarre

> > circumstances.

>

> {mi} can have a plural referent considered as a mass, in which case

> {mi citka lo bakni} can be true, with all of us together eating the cow.

Yes. And {mi citka lo pa bakni} would make it clear that it

is one cow and not several cows or an uncountable amount

of cow.

BTW, Pierre, are you still opposing the definitions because of

the outer-inner quantifiers interplay?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 18:38 GMT

xod:

> I

> can't imagine an example of cows working cooperatively and not as a

> number of individuals.

Cooperative cows:

le sasfoi cu culno lo renono bakni

The meadow is filled with two hundred cows.

None of the cows fills the meadow by itself.

le so'i bakni cu dukse lo ka tilju kei lo nu le karce cu bevri by

The many cows are too heavy for the truck to carry them.

None of the cows is too heavy by itself for the truck to carry it.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 20:03 GMT

On Thu, Jun 03, 2004 at 02:15:19PM -0700, Jorge Llamb?as wrote:

> arj:

> > 1. The new {lo} will have essentially two different meanings with

> > and without quantifiers.

> >

> > Icky and intellectual unfulfilling, but not really a problem.

> > {ku} and {bo} already have different meanings in different

> > contexts.

>

> I don't think the meaning of {lo} changes when you quantify over

> instances. {PA lo broda} is essentially {PA mupli be lo broda}, or {PA

> da poi mupli lo broda}

If this is a strict equivalence, I'd like to see it in the proposal

somewhere.

-Robin



Posted by Anonymous on Fri 04 of June, 2004 20:03 GMT


> On Thu, Jun 03, 2004 at 02:15:19PM -0700, Jorge Llamb?as wrote:

> > I don't think the meaning of {lo} changes when you quantify over

> > instances. {PA lo broda} is essentially {PA mupli be lo broda}, or {PA

> > da poi mupli lo broda}

> If this is a strict equivalence, I'd like to see it in the proposal

> somewhere.

It only works with {mupli} meaning "x1 is an instance of x2".

Unfortunately, that's not quite how the gi'uste defines {mupli}.

According to the gi'uste, mupli means "x1 ckaji x2 gi'e cmima x3".

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 20:03 GMT

On Fri, Jun 04, 2004 at 12:52:31PM -0700, Jorge Llamb?as wrote:

>

> --- Robin Lee Powell wrote:

> > On Thu, Jun 03, 2004 at 02:15:19PM -0700, Jorge Llamb?as wrote:

> > > I don't think the meaning of {lo} changes when you quantify over

> > > instances. {PA lo broda} is essentially {PA mupli be lo broda}, or

> > > {PA da poi mupli lo broda}

> >

> > If this is a strict equivalence, I'd like to see it in the proposal

> > somewhere.

>

> It only works with {mupli} meaning "x1 is an instance of x2".

> Unfortunately, that's not quite how the gi'uste defines {mupli}.

What about 'me' then?

-Robin



Posted by Anonymous on Fri 04 of June, 2004 20:03 GMT


> What about 'me' then?

{me} and {lo} cancel each other out, yes. {lo} converts

selbri to sumti, {me} converts sumti to selbri.

So basically {lo me ko'a} = {ko'a}, {me lo broda} = {broda}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 21:03 GMT

Z> Seems to follow from some readings of {lo}:the most generic absence of unique quantifier would cover the {piPA}s as well as the PAs.

X> bones, hooves, and all?

Pierre Abbat wrote: On Friday 04 June 2004 11:35, xod wrote:

> And it could easily be interpreted as a herd of cows standing in the

> road. Therefore I suggest so'o bakni for cow herd, and tu'o bakni for

> cow goo, and the disuse of loi.

{so'o bakni} is equivalent to {so'o lo bakni} which refers to several cows

considered individually. {loi bakni}, referring to the same cows, considers

them as a mass.

> lo bakni can refer to part of a cow.

Z>I don't see that in the definition of {bakni}.

> Context should *always* imply that you didn't consume the entire cow,

> and you shouldn't need to emphasize that except in the most bizarre

> circumstances.

X>{mi} can have a plural referent considered as a mass, in which case {mi citka

lo bakni} can be true, with all of us together eating the cow.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 21:03 GMT

T> Can individual cows mill around and block a road? Sounds like a herd to me.

S> the game at the moment is to be minimally precise — hence the possibility that {lo bakni} means "a chunk of cow" and {mi} means "all for whom I presume to speak."

xod wrote:

Pierre Abbat wrote:

>On Friday 04 June 2004 11:35, xod wrote:

>

>

>>And it could easily be interpreted as a herd of cows standing in the

>>road. Therefore I suggest so'o bakni for cow herd, and tu'o bakni for

>>cow goo, and the disuse of loi.

>>

>>

>

>{so'o bakni} is equivalent to {so'o lo bakni} which refers to several cows

>considered individually. {loi bakni}, referring to the same cows, considers

>them as a mass.

>

>

T>If loi is never used for cow herds or cow goo, but only in cases where

the cows are acting collectively (doing something as a group that they

could not do individually) then loi has a useful and clear meaning. I

can't imagine an example of cows working cooperatively and not as a

number of individuals. The classic case of this is the 3 men carrying

the piano; a collective task because no one individual can do it, thus

the ability to carry a piano is an emergent property of the collective.

>>lo bakni can refer to part of a cow.

>>

>>

>

>I don't see that in the definition of {bakni}.

>

>

Half a cow is {.5 bakni}. Two cows is {2 bakni}. If {lo bakni} can refer

to the latter, then why not the former?

>>Context should *always* imply that you didn't consume the entire cow,

>>and you shouldn't need to emphasize that except in the most bizarre

>>circumstances.

>>

>>

>

>{mi} can have a plural referent considered as a mass, in which case {mi citka

>lo bakni} can be true, with all of us together eating the cow.

>

>

S>Yes, when mi is referring to a group with the ability to consume a cow,

then you should be clear to distinguish whether they did devour one

animal, or simply sit down to a steak meal.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Fri 04 of June, 2004 21:03 GMT

On Friday 04 June 2004 16:31, John E Clifford wrote:

> Z> Seems to follow from some readings of {lo}:the most generic absence of

> unique quantifier would cover the {piPA}s as well as the PAs.

>

> X> bones, hooves, and all?

lo'e resrvarano cu tai citka, but I did say "at least as much as is edible"

earlier.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 21:03 GMT

Using xorxes' internal quantifiers here (which maybe does away with {loi}).

Jorge Llambías wrote:

xod:

> I

> can't imagine an example of cows working cooperatively and not as a

> number of individuals.

Cooperative cows:

le sasfoi cu culno lo renono bakni

The meadow is filled with two hundred cows.

None of the cows fills the meadow by itself.

le so'i bakni cu dukse lo ka tilju kei lo nu le karce cu bevri by

The many cows are too heavy for the truck to carry them.

None of the cows is too heavy by itself for the truck to carry it.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 21:03 GMT

The examples make {me} seem like a waste of a good and useful; bit of wordspace. I asssume there are examples that makt this word useful enough to justify using up a CV. Where is this meaning spelled out — not on my wordlist.

Jorge Llambías wrote:


> What about 'me' then?

{me} and {lo} cancel each other out, yes. {lo} converts

selbri to sumti, {me} converts sumti to selbri.

So basically {lo me ko'a} = {ko'a}, {me lo broda} = {broda}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Fri 04 of June, 2004 21:05 GMT

Sorry, I never can tell in advance who is operating at what level of precision.

resrvarano? A fu'ivla for a type of snake? Can sets eat?

Pierre Abbat wrote: On Friday 04 June 2004 16:31, John E Clifford wrote:

> Z> Seems to follow from some readings of {lo}:the most generic absence of

> unique quantifier would cover the {piPA}s as well as the PAs.

>

> X> bones, hooves, and all?

lo'e resrvarano cu tai citka, but I did say "at least as much as is edible"

earlier.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 21:15 GMT

On Fri, Jun 04, 2004 at 02:03:19PM -0700, John E Clifford wrote:

> Sorry, I never can tell in advance who is operating at what level of

> precision. resrvarano? A fu'ivla for a type of snake?

A monitor lizard.

http://www.lojban.org/jbovlaste/wiki/taxonomy

BTW, PC, *PLEASE* trim your replies. You have made this thread *much*

larger than it needs to be.

-Robin



Posted by Anonymous on Fri 04 of June, 2004 21:15 GMT

On Friday 04 June 2004 16:54, John E Clifford wrote:

> The examples make {me} seem like a waste of a good and useful; bit of

> wordspace. I asssume there are examples that makt this word useful enough

> to justify using up a CV. Where is this meaning spelled out — not on my

> wordlist.

{me} is also used with cmene (making a type-2 fu'ivla, if the cmevla is

borrowed) and numbers:

mi zbasu lo me me'o bi jgena - I make a figure 8 knot.

ti me la meipl. la sakta - this is a sugar maple.

(For the latter, I'd rather say {.a'orne}, but for a plant whose only name is

in the Waorani language, I'd use a type-1 or type-2.)

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Fri 04 of June, 2004 22:16 GMT

pc:

> The examples make {me} seem like a waste of a good and useful; bit of

> wordspace. I asssume there are examples that makt this word useful enough to

> justify using up a CV. Where is this meaning spelled out — not on my

> wordlist.

CLL talks about it.

Using it with a bare {lo} makes little sense, but it has uses

with any other type of sumti: me KOhA, me la CMENE, me le broda,

me zo , etc.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Too much nesting.


rlpowell Posted by rlpowell on Sat 05 of June, 2004 02:32 GMT posts: 14214

I'm with Pierre WRT the fact that lo quantifiers are different from all the others, actually. I don't think it's a good idea.

It doesn't much matter to me which one makes more sense; it is both an inconsistency and a change to past usage.

We already have other ways of saying "three groups of five"; "ci lo mumei broda" does the right thing, does it not?

Given that, why have the inconsistency? "Three out of some five-some of doctors" may not mean much, but it means something, and it'll be a lot less confusing that way, nevermind the possiblity of invalidating past usage.

Here's an example usage with lo quantifiers being the same as all others:

doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri lo mudri gi'e lo re drata cu kajde fi lo rokci

Not amazingly important, but certainly not completely without use.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 06:45 GMT

On Friday 04 June 2004 22:32, [email protected] wrote:

> I'm with Pierre WRT the fact that lo quantifiers are different from all the

> others, actually. I don't think it's a good idea.

>

> It doesn't much matter to me which one makes more sense; it is both an

> inconsistency and a change to past usage.

>

> We already have other ways of saying "three groups of five"; "ci lo mumei

> broda" does the right thing, does it not?

"ci lo broda mumei", or to avoid tanru, "ci lo mumei be fi lo broda".

> Given that, why have the inconsistency? "Three out of some five-some of

> doctors" may not mean much, but it means something, and it'll be a

> lot less confusing that way, nevermind the possiblity of invalidating past

> usage.

>

> Here's an example usage with lo quantifiers being the same as all others:

>

> doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri lo

> mudri gi'e lo re drata cu kajde fi lo rokci

Sounds good to me. Are they building a fort, or what?

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Sat 05 of June, 2004 06:45 GMT

Robin:

> It doesn't much matter to me which one makes more sense; it is both an

> inconsistency and a change to past usage.

What past usage?

> We already have other ways of saying "three groups of five"; "ci lo mumei

> broda" does the right thing, does it not?

It's a tanru. It gets the message across, but we want to have more

precise ways of talking too.

> Given that, why have the inconsistency? "Three out of some five-some of

> doctors" may not mean much, but it means something, and it'll be a

> lot less confusing that way, nevermind the possiblity of invalidating past

> usage.

That is a very complex formula in quantifier terms. "There is some

group of five such that three members of that group ..."

How is the group of five determined, given that it is not an

in-mind selection of the speaker and the claim only

holds for three members? Where do the other two come from?

> Here's an example usage with lo quantifiers being the same as all others:

>

> doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri lo mudri

> gi'e lo re drata cu kajde fi lo rokci

(Both {gi'e} should be {ije}.) Shouldn't those be {le}?

In any case, we have:

PA lo broda = PA mupli be lo broda

PA le broda = PA cmima be le broda

Having

PA lo broda = PA cmima be lo mupli be lo broda

does not strike me as more regular.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 06:45 GMT

On Fri, Jun 04, 2004 at 10:53:58PM -0400, Pierre Abbat wrote:

> On Friday 04 June 2004 22:32, [email protected] wrote:

> > Here's an example usage with lo quantifiers being the same as all

> > others:

> >

> > doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri

> > lo mudri gi'e lo re drata cu kajde fi lo rokci

>

> Sounds good to me. Are they building a fort, or what?

Just moving logs; watching for rocks so the carriers don't trip.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 06:45 GMT

On Fri, Jun 04, 2004 at 08:10:03PM -0700, Jorge Llamb?as wrote:

>

> Robin:

> > It doesn't much matter to me which one makes more sense; it is both

> > an inconsistency and a change to past usage.

>

> What past usage?

>From Le Petit Prince 2?:

mi krefu finti seva'u py pa lo re po'o pixra poi mi kakne

I find it *profoundly* unlikely that that's meant to mean anything other

than "one of the two pictures which I am able to draw". In fact,

that's a *perfect* usage example, IMO.

Do you wish me to find more?

> > We already have other ways of saying "three groups of five"; "ci lo

> > mumei broda" does the right thing, does it not?

>

> It's a tanru. It gets the message across, but we want to have more

> precise ways of talking too.

Are there none?

> > Given that, why have the inconsistency? "Three out of some

> > five-some of doctors" may not mean much, but it means

> > something, and it'll be a lot less confusing that way, nevermind

> > the possiblity of invalidating past usage.

>

> That is a very complex formula in quantifier terms. "There is some

> group of five such that three members of that group ..."

Doesn't seem particularily hard to me.

> How is the group of five determined, given that it is not an in-mind

> selection of the speaker and the claim only holds for three members?

> Where do the other two come from?

However you like; this is generic lo, after all.

> > Here's an example usage with lo quantifiers being the same as all

> > others:

> >

> > doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri

> > lo mudri gi'e lo re drata cu kajde fi lo rokci

>

> (Both {gi'e} should be {ije}.)

Yes.

> Shouldn't those be {le}?

No, because the speaker is referring to all groups in question and cares

not a whit how they are divided.

> In any case, we have:

>

> PA lo broda = PA mupli be lo broda

> PA le broda = PA cmima be le broda

>

> Having

>

> PA lo broda = PA cmima be lo mupli be lo broda

>

> does not strike me as more regular.

As none of that is part of the definition, I don't see how it's

relevant.

The regularity is in usage. The outer quantifier works the same for

  • all* gadri *except* lo.

mu le pa broda

and

mu lo pa broda

are talking about completely different numbers of things. Better yet,

the former is invalid but the latter is not. That's going to be very

confusing.

Quite frankly, I'd rather that all quantifiers couted groups, but that

would break past usage much more badly.

-Robin


Robin's Long Post


rlpowell Posted by rlpowell on Sat 05 of June, 2004 06:55 GMT posts: 14214

(Dammit, I just lost everything I had done. Grr. mad Restarting).

This is my requisite long post, in which I show myself to be the only

person in the BPFK that actually reads the Lojban carefully and is

capable of copy-editing. If I sound annoyed, it's because I am, but only

a little bit.

One of these suggestions is a joke. See if you can spot it.

At the beginning of this post, I was politely phrasing things as

requests. Eventually this got tedious; please insert "please" and "if

you don't mind" anywhere it will make you feel better.

  • "It converts a selbri, selecting its first argument, into a sumti.".

"Selecting" is rather opaque to me. Maybe you could add something like

"In other words, lo broda is anything that could fit the first argument

of broda". Also see the next point. This applies to, umm, all of the

definitions, as does the next point.

  • "that satisfies the predicate." should be "that satisifies the first

argument of the predicate". HOWEVER, please read the next few

points before doing anything about this one.

  • In loi: "The resulting expression refers to a group of individuals

that satisfy the selbri and which satisfy the predicate for which the

sumti is an argument collectively." This is different from every other

definition for no apparent reason. I actually like the extra

specificity, but please either repeat it everywhere or drop it.

  • OK, it may not actually be true that the others are less

specific; it may be that "the predicate" in "and which the speaker

describes with the predicate" is intended to refer to the whole

bridi, not just the selbri. In which case, I very strongly

request that you add the "for which the sumti is an argument" thing

everywhere, because by "the predicate", I thought you meant the

predicate represented by the selbri.

  • This point may very well abrogate the previous point; I don't know

which you meant "the predicate" to mean, but please pick one and

make it clear.

  • "Children should always" . "always" does not occur in the Lojban; fix

one or the other please.

  • I do not know what "viska pa'o lo tanxe" means, but I am certain it

does not mean " through the walls of boxes"; add bitmu or

something please.

  • "lo ctuca cu fendi lo selctu mu lo vo tadni" . the first two lo are

most definately "le" if you want it to match the English. You should

probably quantify ctuca as well.

  • "lo kucysni" . given the numerical precision of the rest, you should

quantify this.

  • Was it really necessary to make a lujvo out of "Soviet Cow?"

biggrin

  • The unicorn one:
  • "gi'e catlu lo ka" — did you mean "se catlu"? No, not even that

works. You want "And can be seen to have"; "se catlu simsa" would

work I think.

  • Quantify jirna, please; it just sounds silly otherwise.
  • Forehead is "mebri"
  • What does "describes with the predicate" mean? If you are maintaining

the non-veridicality of "le", please be specific about that. Perhaps

"chooses to describe with the first place of the predicate, whether it

objectively applies or not". But see the first few points; I may

have had the wrong predicate here.

  • Similarily, please include at least one blatantly non-veridical

example for "le". Ideally, have one for each non-veridical gadri, but I

won't bitch if you don't wont to bother with ones other than "le".

  • "An inner quantifier can be used in the case of a selbri to indicate

the cardinality of the group." Explain to me why this doesn't work with

cmevla? "ci la pano hels.angels. pu darxi mi doi pulji". checks

Wow, that's a grammar change! OK, that's reason enough. Kinda sad

though. We do have the ability to make grammar changes; your call

though, I'm not set on one or the other. However, I suddenly

understand your dislike of the seperate LA selma'o.

  • "ca jbena" seems insufficient; "puzi jbena" or "cazi jbena" would be

nicer I think.

  • The emphasis in "FACE-DOWN" is not mirrored in the Lojban.
  • lei brazo/dotco prenu, please.
  • "The resulting expression refers to the typical individual or

group that satisfies the predicate.". Don't you mean a typical

individual or group? If not, more explanation please. Same with

lo'e.

  • For completeness, we really need set and mass versions of lo'e and

le'e.

  • Umm, isn't it the outer quantifier of lo that is different?
  • I would really like something more in there about constants. A

link to a really good expository URL would be nice. Even something

as simple as "A constant is something that has no quantity from a

predicate logic perspective" or something would help.

  • Please make it clear the tu'o thing is provisional until the BPFK

gets there.

  • In the "tavla fi le tutra pe le terdi" example: "mokla tirxe"

should be "moklu trixe".

  • "In the mountains there is no food." s/is/will be/
  • "so'omoi" — probably meant to be mei.

Translations.


rlpowell Posted by rlpowell on Sat 05 of June, 2004 07:10 GMT posts: 14214

Every time I go through a section, I translate all the Lojban independently of the English, to check for errors. Just for giggles, here's what I got this time. These are (deliberately) very literalistic.

ei lo verba cu mutce fraxu lo makcu prenu

Obligation: Children should muchly forgive mature persons/non-children.

ku'i uinai mi na viska lo lanme pa'o lo tanxe

i ju'ocu'i mi milxe simsa lo makcu prenu

However, sadly, I don't see sheep passing through boxes. Maybe I am

mildly similar to mature persons.

ca lo nicte lo cinfo cu kalte lo cidja

During night-time, lions hunt food.

lo pa pixra cu se vamji lo ki'o valsi

A one-some of pictures (i.e. any single picture) has value a

thousand-some (i.e. any thousand) words.

de'i li 1960 lo pare sovda cu fepni li 42

In 1960, 12 eggs were measured in cents numbering 42.

lo ctuca cu fendi lo selctu mu lo vo tadni

Teachers (un-numbered, could be one) divide those taught into five

foursomes of students.

lo bidjylinsi pe lo ze seldri cu se pagbu

ze lo ze bidju e ji'a ci lo pa bidju e lo kucysni

Necklaces of seven-somes of things causing sadness have parts: seven

seven-somes of been and also three one-somes of beeds and crosses

(un-numbered, could be one).

o'i mu (lo) xagji sofybakni cu zvati le purdi

Beware: five hungry Soviet cows are in the garden.

lo sanli darxi bo dakli cu culno lo djacu onai lo canre to lo djacu

cu pukmau ki'u lo nu slilu tolcando toi gi'e bunda li ji'i 270

Standing hit-sacks are full of water XOR sand (water is more pleasant

because of the event of oscillating non-idleness (motion) ) and weigh

about 270 pounds.

lo pavyseljirna cu ranmi danlu gi'e catlu lo ka ge ce'u xirma

gi lo jirna cu cpana lo sedycra be ce'u

Unicorns are mythical animals and examine se catlu meant? the property

of both it's a horse and horns (un-numbered, could be one) are upon the

front-head I suggest mebri of it.

- -------

le palta ba'o porpi i ma gasnu i xu le gerku cu go'i

The plate has been broken. Who did? Is it true that the dog did?

ko punji le sicni lo porsi be lo vamrai bi'o lo vamtolrai

Put the coin in the sequence ordered from most expensive to least

expensive.

ci le bi ctuca cu ninmu

Three of the eight teachers are women.

- -------

ma'i la midju terdi la sadam na sai me la sauron

In the reference frame of Middle Earth, Saddam is very much not Sauron.

- -------

ma cnano lo ka makau junta ce'u kei lo'i cifnu poi ca jbena

What is normal in what weight it is of the set of babies now born?

- -------

ro le verba pu cuxna pa karda le'i cnita selcra

All of the children chose one card from this set of face-down.

- -------

doi turni do so'i da na fadni la'i kenedis ma'i lo jecra'a

Oh ruler, you are in many ways not ordinary among the set of things

named "kenedis" in reference to polity-relevant-things.

- -------

loi litru ti jmaji lo ro pagbu be le terdi

A mass of travellers here gathers from every part of the earth.

- -------

lei brazo cu jinga fi lei dotco la kabri

The mass of Brazillian things won agains the mass of German things in a

competition called Cup.

- -------

ta melbi pixra lai simpson

That is a beautiful picture of the mass of things named Simpson.

- -------

lo'e glipre cu xabju le fi'ortu'a na.e le gligugde

The typical Englishman dwells in African, *not*, and in England.

- -------

le'e xelso merko cu gusta ponse

Stereotypical Greek Americans are restaurant type-of owners.

- -------

le nanmu cu se snuti ija'ebo lo tu'o gerku cu kuspe le klaji

The man did something accidental and therefore a mass of doggy bits

spanned the road.

bilga lenu jdice lenu roroi pilno lo mokla tirxe

(to zoigy. velar gy. toi) jonai crane (to zoigy.

alveolar gy. toi)

It is obliged that it is decided that every time using mouth behind-

("Velar") -XOR-front ("alveolar")

le cmana lo cidja ba claxu

The mountain, food will be without.

Yoda I am, hmm, yeees!

le dargu pe lo xamgu bangu cu kargu

The road of good languages is costly.

la jyryr. tolkien. cu te cukta la djine turni (to la'o

gy Lord of the Rings gy toi) .e le so'omoi be lo

xanri munje lisri ca le lampru na'acto

JRR Tolkien wrote Ring Lord ("Lord Of The Rings") and the several-th of

imaginary-universe storiies during the previous century.



Posted by Anonymous on Sat 05 of June, 2004 12:13 GMT

[email protected] wrote:

>* Please make it clear the tu'o thing is provisional until the BPFK

>gets there.

>

>

tu'o is a "best practice"; it is a usage that doesn't affect the

official meaning of tu'o, and therefore never goes before the BPFK.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Sat 05 of June, 2004 12:13 GMT

On Sat, Jun 05, 2004 at 04:28:45AM -0400, xod wrote:

> [email protected] wrote:

>

> >* Please make it clear the tu'o thing is provisional until the BPFK

> >gets there.

>

> tu'o is a "best practice"; it is a usage that doesn't affect the

> official meaning of tu'o, and therefore never goes before the BPFK.

That's not what xorxes has been saying.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 12:13 GMT

On Saturday 05 June 2004 02:55, [email protected] wrote:

> * Was it really necessary to make a lujvo out of "Soviet Cow?"

> biggrin

It was necessary to have a 'y' in the sentence to make it a pangram, and the

most obvious way to get one was to make a lujvo.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Sat 05 of June, 2004 14:22 GMT

Robin Lee Powell:

> From Le Petit Prince 2?:

:-)

> mi krefu finti seva'u py pa lo re po'o pixra poi mi kakne

Ok, that's one example. It doesn't make me change my mind, though.

> Do you wish me to find more?

If it's not too much trouble, it would be great.

> > > We already have other ways of saying "three groups of five"; "ci lo

> > > mumei broda" does the right thing, does it not?

> >

> > It's a tanru. It gets the message across, but we want to have more

> > precise ways of talking too.

>

> Are there none?

Sure, but much wordier.

We'd also lose the quantification over fractions {PA lo piPA broda}.

> The regularity is in usage. The outer quantifier works the same for

> *all* gadri *except* lo.

  • all* is le and la, right?

It certainly doesn't work like that for loi, lei, lai, lo'i, le'i,

la'i, lo'e and le'e.

> Quite frankly, I'd rather that all quantifiers couted groups, but that

> would break past usage much more badly.

If you have more than one group in mind, you can still manage with {le}:

le ci lo mu broda

The three five-brodas

{le} points to a single thing you have in mind (in that example the

single thing is the group of three five-brodas). You can quantify

over members of the thing, but not over instances.

{lo} points to the predicate that must be satsfied. The natural thing

to quantify over are the things that satisfy the predicate, not the

members of a group that satisfies the predicate.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 14:22 GMT

Robin:

> One of these suggestions is a joke. See if you can spot it.

>

> * "It converts a selbri, selecting its first argument, into a sumti.".

> "Selecting" is rather opaque to me. Maybe you could add something like

> "In other words, lo broda is anything that could fit the first argument

> of broda".

I added something like that, please check if it's clear.

....

> ** This point may very well abrogate the previous point; I don't know

> which you meant "the predicate" to mean, but please pick one and

> make it clear.

Yes, I was using "predicate" for two different things. I now use "selbri"

in all cases for the transformed selbri, and "predicate for which the sumti

is an argument" for the external one. Please check if it's clear.

> * "Children should always" . "always" does not occur in the Lojban; fix

> one or the other please.

This is what's going on: The Lojban is a translation from the French.

The English is also a translation from the French, by someone other

than me. In French, there is no explicit "always". Maybe the

translator thought that "always" made the intent of the original

more clear in English. Anyway, I removed it.

> * I do not know what "viska pa'o lo tanxe" means, but I am certain it

> does not mean " through the walls of boxes"; add bitmu or

> something please.

Same as before, no mention of walls in French. I wouldn't mention walls

in Spanish either, but I will add the walls to make it more

English-friendly. {pa'o lo bitmu be fo lo tanxe} then.

> * "lo ctuca cu fendi lo selctu mu lo vo tadni" . the first two lo are

> most definately "le" if you want it to match the English. You should

> probably quantify ctuca as well.

How would you translate the Lojban into normal English? (The context

is a set of instructions for conducting a Lesson.)

> * "lo kucysni" . given the numerical precision of the rest, you should

> quantify this.

Only if you think that I should also change to "one Crucifix" in English.

But I changed {ci lo pa bidju} to {ci bidju}, which is clearly enough.

> * The unicorn one:

> * "gi'e catlu lo ka" — did you mean "se catlu"? No, not even that

> works. You want "And can be seen to have"; "se catlu simsa" would

> work I think.

I meant {simlu}! Thanks for catching that one!

> * Quantify jirna, please; it just sounds silly otherwise.

Ok, but I'm not sure why it sound silly.

> * Forehead is "mebri"

Right.

> * Similarily, please include at least one blatantly non-veridical

> example for "le". Ideally, have one for each non-veridical gadri, but I

> won't bitch if you don't wont to bother with ones other than "le".

Any suggestions for something natural? I don't want to give the

impression that non-veridicality is used in more that .1% of cases.

> * "An inner quantifier can be used in the case of a selbri to indicate

> the cardinality of the group." Explain to me why this doesn't work with

> cmevla?

Because the grammar is overrestrictive. CMENE should be in BRIVLA.

> "ci la pano hels.angels. pu darxi mi doi pulji". checks

> Wow, that's a grammar change! OK, that's reason enough. Kinda sad

> though. We do have the ability to make grammar changes; your call

> though, I'm not set on one or the other. However, I suddenly

> understand your dislike of the seperate LA selma'o.

Is this the appropriate place and time to propose moving CMENE

into BRIVLA?

> * "ca jbena" seems insufficient; "puzi jbena" or "cazi jbena" would be

> nicer I think.

Ok {cazi}.

(jbena is really wrong for this, because it has a time place,

so it doesn't make much sense to use a tense with it, but

I'm not going to use zi'o.)

> * The emphasis in "FACE-DOWN" is not mirrored in the Lojban.

That's how it was written where I found it. It didn't strike me

as emphatic though. I'm lower-casing it.

> * lei brazo/dotco prenu, please.

Why?

> * "The resulting expression refers to the typical individual or

> group that satisfies the predicate.". Don't you mean a typical

> individual or group? If not, more explanation please. Same with

> lo'e.

I think it's "the" typical, it's a single abstraction. See if you like

the new version.

> * For completeness, we really need set and mass versions of lo'e and

> le'e.

And also name versions.

> * Umm, isn't it the outer quantifier of lo that is different?

The inner of lo becomes the cadinality of a generic group instead of

being the cardinality of the group of all existing brodas.

The outer is adjusted accordingly, but {PA lo broda} retains the

same meaning.

> * I would really like something more in there about constants. A

> link to a really good expository URL would be nice. Even something

> as simple as "A constant is something that has no quantity from a

> predicate logic perspective" or something would help.

A constant is something that always keeps the same referent. {lo broda}

always refers to brodas. In {mu da poi broda zo'u da brode}, "da" is

a variable, because it takes values from the set of all things that

brodas. Anything with a quantifier in front takes values from the set

of things over which the quantifier runs.

> * Please make it clear the tu'o thing is provisional until the BPFK

> gets there.

I changed "can" to "could".

> * In the "tavla fi le tutra pe le terdi" example: "mokla tirxe"

> should be "moklu trixe".

I'm quoting.

> * "In the mountains there is no food." s/is/will be/

Helselm's words, not mine.

> * "so'omoi" — probably meant to be mei.

Probably, but I'm just quoting.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 16:53 GMT

Robin Lee Powell wrote:

>On Sat, Jun 05, 2004 at 04:28:45AM -0400, xod wrote:

>

>

>>[email protected] wrote:

>>

>>

>>

>>>* Please make it clear the tu'o thing is provisional until the BPFK

>>>gets there.

>>>

>>>

>>tu'o is a "best practice"; it is a usage that doesn't affect the

>>official meaning of tu'o, and therefore never goes before the BPFK.

>>

>>

>

>That's not what xorxes has been saying.

>

>

Which has he been saying?

--

Motorists honked in celebration in this Ramadi as news spread of the

assassination of the president of the Iraqi Governing Council Ezzidin Salim

Monday. "The GC is nothing," one man shouted. "They are not the Governing

Council. They are the Prostitution Council."



Posted by Anonymous on Sat 05 of June, 2004 19:11 GMT

On Sat, Jun 05, 2004 at 06:33:49AM -0400, Pierre Abbat wrote:

> On Saturday 05 June 2004 02:55, [email protected] wrote:

> > * Was it really necessary to make a lujvo out of "Soviet Cow?"

> > biggrin

>

> It was necessary to have a 'y' in the sentence to make it a pangram,

> and the most obvious way to get one was to make a lujvo.

Yeah, didn't realize it was a pangram until xod pointed it out.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 19:16 GMT

> >>>* Please make it clear the tu'o thing is provisional until the BPFK

> >>>gets there.

> >>>

> >>tu'o is a "best practice"; it is a usage that doesn't affect the

> >>official meaning of tu'o, and therefore never goes before the BPFK.

> >

> >That's not what xorxes has been saying.

>

> Which has he been saying?

Quoting from a conversation between him and arj:

> My objections to the current proposal that are as yet unresolved:

>

> 3. {tu'o} as an inner quantifier is, as I understand it, either a

> special

> case that magically turns {lo} into a generic mass article, or else is

> intended to be a part of the general quantifier system.

{tu'o} is not mentioned in the definition of {lo}, not even

in the examples. Its meaning as an inner quantifier has to be

proposed and voted on when defining {tu'o}, that's why I only

mentioned it under "notes".

> If the latter, it does not appear to fit stringently into the system.

> It appears to be one of these clever tricks that immediately make

> sense, but does not really hold when people try to think inside the

> system, instead of standing on the outside1.

If you don't think it fits into the system, vote against it when

it is proposed in the definition of {tu'o}.

In resonse to PC (I think):

... I didn't want to include {tu'o}

in the definition of {lo} because it is rather marginal. I doubt

it will be used much if at all. Substances don't really

need special marking in most cases. Since this is really

about {tu'o} more than about {lo}, it should be treated in

the definition of {tu'o}.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 19:35 GMT

On Sat, Jun 05, 2004 at 06:13:12AM -0700, Jorge Llamb?as wrote:

> Robin Lee Powell:

> > From Le Petit Prince 2?:

>

> :-)

Yes, that amused me as well.

> > Do you wish me to find more?

>

> If it's not too much trouble, it would be great.

OK, here's the thing:

It seems to me that this changes *all* usage of "PA lo", because

suddenly when you said "pa lo broda" and meant "One broda", you don't

  • get* one broda anymore, you get one *group* of broda, which is very,

very different.

" If I say "pa lo re nanmu", I make a much stronger claim. I am of

course selecting one member from the set of things which really are men

to discuss; I am also stating that this set is enumerated as having two

members. "

http://www.lojban.org/files/draft-textbook/lesson19

Which is, of course, the most important point: This is a drastict

change from current teaching materials. Dropping implicit quantifiers

seems, to me, much less drastic.

su'o pa lo prenu cu prami do

At least one person loves you.

http://www.lojban.org/files/brochures/lesson4.html

In your version, this means "At least one group of people loves you",

does it not?

Fully expanded, su'o pa lo pa broda: at least one out of the one thing

in the world which can be described (veridically) as the x1 of broda.

Since brivla typically describe more than one thing, it's hard to find

examples, so let's settle for lo pa cevni be le xebro. (There is not

only one God for all religions, but there is only one God for that

religion.)

http://www.lojban.org/tiki/only

I can't decied if the stuff on

http://www.lojban.org/tiki/sumti+qualifiers agrees

with you or not.

19 May 2003 14:23:19 ci re'u ca pa lo cacra

http://www.lojban.org/resources/irclog/lojban/2003_05_20-02_22.txt

Again, your version would be "Three times in one *group* of hours".

That was just from searching on 'site:www.lojban.org "pa lo"', and I

ommited most of the ones that used only an outer quantifier.

I immediately see one example from searching on "re lo".

> > The regularity is in usage. The outer quantifier works the same for

> > *all* gadri *except* lo.

>

> *all* is le and la, right?

> It certainly doesn't work like that for loi, lei, lai, lo'i, le'i,

> la'i, lo'e and le'e.

You have very few examples of quantification of those ones. However,

those all say "An outer quantifier can be used to indicate a subset of

that cardinality ", or "subgroup" instead of subset. la and le say " An

outer quantifier can be used to quantify over members of the group."

Ignoring lo'e and le'e, of course.

If those two are different, I don't understand one or the other.

The only example of quantification of these articles is "ro le verba",

which helps very little.

If "indicate a subset of that cardinality" and "quantify over members of

the group" mean substantially different things, then on behalf of slow

people everywhere I request more verbosity. "In other words, ..." would

be nice. More examlpes would be nice too.

> > Quite frankly, I'd rather that all quantifiers couted groups, but

> > that would break past usage much more badly.

>

> If you have more than one group in mind, you can still manage with

> {le}:

>

> le ci lo mu broda

> The three five-brodas

I had no idea that was legal.

> {le} points to a single thing you have in mind (in that example the

> single thing is the group of three five-brodas). You can quantify over

> members of the thing, but not over instances.

>

> {lo} points to the predicate that must be satsfied. The natural thing

> to quantify over are the things that satisfy the predicate, not the

> members of a group that satisfies the predicate.

I understand your point; it's the change in usage and inconsistency that

bother me.

I will probably not vote No just for this reason, however.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 20:11 GMT

J> I would argue that {ci lo broda mumei} is better, we want a brodaish fivesome, not a fivesomeish broda. As xorxes notes, all of these are tanru and so open to other interpretations — so go for a lujvo.

[email protected] wrote:Too much nesting.

I'm with Pierre WRT the fact that lo quantifiers are different from all the others, actually. I don't think it's a good idea.

It doesn't much matter to me which one makes more sense; it is both an inconsistency and a change to past usage.

J>We already have other ways of saying "three groups of five"; "ci lo mumei broda" does the right thing, does it not?

Given that, why have the inconsistency? "Three out of some five-some of doctors" may not mean much, but it means something, and it'll be a lot less confusing that way, nevermind the possiblity of invalidating past usage.

Here's an example usage with lo quantifiers being the same as all others:

doi lo re no prenu ko fendi ko lo mumei gi'e ci lo mu prenu cu bevri lo mudri gi'e lo re drata cu kajde fi lo rokci

Not amazingly important, but certainly not completely without use.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 20:11 GMT

On Sat, Jun 05, 2004 at 07:19:09AM -0700, Jorge Llamb?as wrote:

>

> Robin:

> > One of these suggestions is a joke. See if you can spot it.

> >

Did you spot it?

> > * "It converts a selbri, selecting its first argument, into a

> > sumti.". "Selecting" is rather opaque to me. Maybe you could add

> > something like "In other words, lo broda is anything that could fit

> > the first argument of broda".

>

> I added something like that, please check if it's clear.

>

> ...

> > ** This point may very well abrogate the previous point; I don't know

> > which you meant "the predicate" to mean, but please pick one and

> > make it clear.

>

> Yes, I was using "predicate" for two different things. I now use "selbri"

> in all cases for the transformed selbri, and "predicate for which the sumti

> is an argument" for the external one. Please check if it's clear.

"The resulting expression refers specifically to an individual or group

that the speaker has in mind and which the speaker describes with the

selbri." "describes as fitting the first argument of the selbri",

please.

> > * "Children should always" . "always" does not occur in the Lojban;

> > fix one or the other please.

>

> This is what's going on: The Lojban is a translation from the French.

> The English is also a translation from the French, by someone other

> than me. In French, there is no explicit "always". Maybe the

> translator thought that "always" made the intent of the original more

> clear in English.

OK, fair enough.

> Anyway, I removed it.

Thanks.

> > * I do not know what "viska pa'o lo tanxe" means, but I am certain

> > it does not mean " through the walls of boxes"; add bitmu or

> > something please.

>

> Same as before, no mention of walls in French. I wouldn't mention

> walls in Spanish either, but I will add the walls to make it more

> English-friendly. {pa'o lo bitmu be fo lo tanxe} then.

Thanks. I'm still not sure I know what the Lojban means, but it's

certainly close enough.

> > * "lo ctuca cu fendi lo selctu mu lo vo tadni" . the first two lo

> > are most definately "le" if you want it to match the English. You

> > should probably quantify ctuca as well.

>

> How would you translate the Lojban into normal English? (The context

> is a set of instructions for conducting a Lesson.)

The way I translated it (see my other post) is "Teachers (un-numbered,

could be one) divide those taught into five foursomes of students."

> > * "lo kucysni" . given the numerical precision of the rest, you

> > should quantify this.

>

> Only if you think that I should also change to "one Crucifix" in

> English.

"a crucifix" *does* mean "one crucifix", at least in my dialect.

> But I changed {ci lo pa bidju} to {ci bidju}, which is clearly enough.

Actually, that's not clear to me at all; see the other thread.

> > * The unicorn one:

> > * "gi'e catlu lo ka" — did you mean "se catlu"? No, not even that

> > works. You want "And can be seen to have"; "se catlu simsa" would

> > work I think.

>

> I meant {simlu}! Thanks for catching that one!

Cool.

> > * Quantify jirna, please; it just sounds silly otherwise.

>

> Ok, but I'm not sure why it sound silly.

Because you have a pavyseljirna with some number of horns unspecified.

> > * Similarily, please include at least one blatantly non-veridical

> > example for "le". Ideally, have one for each non-veridical gadri,

> > but I won't bitch if you don't wont to bother with ones other than

> > "le".

>

> Any suggestions for something natural? I don't want to give the

> impression that non-veridicality is used in more that .1% of cases.

Erk. "le ta ninmu cu mutce melbi .iku'i ca'a nanmu gi'e nelci lo nu

ninmu dasni" is the first one that comes to mind.

Gender-queer-positive people would have a field day with "le", I

suspect.

> > * "An inner quantifier can be used in the case of a selbri to

> > indicate the cardinality of the group." Explain to me why this

> > doesn't work with cmevla?

>

> Because the grammar is overrestrictive. CMENE should be in BRIVLA.

I see that.

> > "ci la pano hels.angels. pu darxi mi doi pulji". checks Wow,

> > that's a grammar change! OK, that's reason enough. Kinda sad

> > though. We do have the ability to make grammar changes; your call

> > though, I'm not set on one or the other. However, I suddenly

> > understand your dislike of the seperate LA selma'o.

>

> Is this the appropriate place and time to propose moving CMENE into

> BRIVLA?

mumble, mumble It's up to you. I can't think of a *better* time,

though. If you do so, I will *immediately* call an extension to voting,

for obvious reasons.

> > * lei brazo/dotco prenu, please.

>

> Why?

Because otherwise we could be talking about Brazillian versus German

sausages, and who one for being placed in a very expensive bowl called

The Cup.

> > * "The resulting expression refers to the typical individual or

> > group that satisfies the predicate.". Don't you mean a typical

> > individual or group? If not, more explanation please. Same with

> > lo'e.

>

> I think it's "the" typical, it's a single abstraction. See if you like

> the new version.

"typically satisfy also the predicate": swap "satisfy" and "also". That

works.

> > * Umm, isn't it the outer quantifier of lo that is different?

>

> The inner of lo becomes the cadinality of a generic group instead of

> being the cardinality of the group of all existing brodas.

>

> The outer is adjusted accordingly, but {PA lo broda} retains the same

> meaning.

I don't see how that's possible. Before, "pa lo broda" meant *exactly*

one broda. Now it means one *group* of broda, of indeterminate size.

That is a massive change, unless I'm missing something.

None of your examples use an outer quantifier by itself, by the way.

Might want to fix that.

> > * I would really like something more in there about constants.

> > A link to a really good expository URL would be nice. Even

> > something as simple as "A constant is something that has no quantity

> > from a predicate logic perspective" or something would help.

>

> A constant is something that always keeps the same referent. {lo

> broda} always refers to brodas. In {mu da poi broda zo'u da brode},

> "da" is a variable, because it takes values from the set of all things

> that brodas. Anything with a quantifier in front takes values from the

> set of things over which the quantifier runs.

Don't tell me; tell the notes. :-)

> > * Please make it clear the tu'o thing is provisional until the BPFK

> > gets there.

>

> I changed "can" to "could".

OK.

-Robin


Attempting to increase clarity.


[[user1|admin]] Posted by admin on Sat 05 of June, 2004 20:15 GMT posts: 208

OK, clarifying my stance on the lo quantifier thing.

It's not the inner quantifier change I'm worried about. The two or three times that someone has really *wanted* to quantify the entire universe of objects can go stuff themselves, although I *am* curious as to how to say "One of the pictures I can draw, of which there are only two in the universe" in the new lo.

What I'm worried about is the change that this effectively makes to the behaviour of the *outer* quantifier of lo. "An outer quantifier can be used to quantify over instances of the generic individual or group.": "or group" is what I'm worried about. "pa lo broda" used to me *exactly* one broda; now it can mean one *group* of broda, unless I'm missing something.

This is a major and drastic change to past usage, and I cannot support it.

I doubt that fixing it is difficult, but it needs to be addressed.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 20:59 GMT

L> A clear variation on the famouse "Juno was a man."

K> but then how keep the quantifier from becoming part of the name — and how distinguish from cases where it is part of the nme "Three Cows Came> {la ci bakni. cu klama} Three cows or Three-Cows?

J> the claim that {lo broda} is a constant because it always refers to broda is at least contentious — and to a logician just blatantly false. Like a variable it ranges over broda; we just don't know how wide the range is.

Jorge Llambías wrote:

Robin:

> * Similarily, please include at least one blatantly non-veridical

> example for "le". Ideally, have one for each non-veridical gadri, but I

> won't bitch if you don't wont to bother with ones other than "le".

L>Any suggestions for something natural? I don't want to give the

impression that non-veridicality is used in more that .1% of cases.

> * "An inner quantifier can be used in the case of a selbri to indicate

> the cardinality of the group." Explain to me why this doesn't work with

> cmevla?

K>Because the grammar is overrestrictive. CMENE should be in BRIVLA.

> "ci la pano hels.angels. pu darxi mi doi pulji". checks

> Wow, that's a grammar change! OK, that's reason enough. Kinda sad

> though. We do have the ability to make grammar changes; your call

> though, I'm not set on one or the other. However, I suddenly

> understand your dislike of the seperate LA selma'o.

Is this the appropriate place and time to propose moving CMENE

into BRIVLA?

> * I would really like something more in there about constants. A

> link to a really good expository URL would be nice. Even something

> as simple as "A constant is something that has no quantity from a

> predicate logic perspective" or something would help.

J>A constant is something that always keeps the same referent. {lo broda}

always refers to brodas. In {mu da poi broda zo'u da brode}, "da" is

a variable, because it takes values from the set of all things that

brodas. Anything with a quantifier in front takes values from the set

of things over which the quantifier runs.

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 20:59 GMT

If you don't start trimming your replies, I'm going to lock you out of

the boards, PC.

> K>Because the grammar is overrestrictive. CMENE should be in BRIVLA.

>

> but then how keep the quantifier from becoming part of the name --

> and how distinguish from cases where it is part of the nme "Three Cows

> Came> {la ci bakni. cu klama} Three cows or Three-Cows?

There's a difference?

-Robin



Posted by Anonymous on Sat 05 of June, 2004 20:59 GMT

Robin:

> On Sat, Jun 05, 2004 at 07:19:09AM -0700, Jorge Llamb?as wrote:

> > Robin:

> > > One of these suggestions is a joke. See if you can spot it.

> Did you spot it?

Yes. Did you think my response to it was serious? :-)

> "The resulting expression refers specifically to an individual or group

> that the speaker has in mind and which the speaker describes with the

> selbri." "describes as fitting the first argument of the selbri",

> please.

OK.

> > > * "lo ctuca cu fendi lo selctu mu lo vo tadni" . the first two lo

> > > are most definately "le" if you want it to match the English. You

> > > should probably quantify ctuca as well.

> >

> > How would you translate the Lojban into normal English? (The context

> > is a set of instructions for conducting a Lesson.)

>

> The way I translated it (see my other post) is "Teachers (un-numbered,

> could be one) divide those taught into five foursomes of students."

I mean into normal English. The Lojban sounds normal as it is.

> > Any suggestions for something natural? I don't want to give the

> > impression that non-veridicality is used in more that .1% of cases.

>

> Erk. "le ta ninmu cu mutce melbi .iku'i ca'a nanmu gi'e nelci lo nu

> ninmu dasni" is the first one that comes to mind.

Added. (With {va} instead of {ta}.)

> Gender-queer-positive people would have a field day with "le", I

> suspect.

Many of them would claim they are ca'a ninmu, not that

they are using the selbri non-veridically.

> > Is this the appropriate place and time to propose moving CMENE into

> > BRIVLA?

>

> mumble, mumble It's up to you. I can't think of a *better* time,

> though. If you do so, I will *immediately* call an extension to voting,

> for obvious reasons.

I won't then. We can always fix the definition of {la} later if that is

ever changed.

> > > * lei brazo/dotco prenu, please.

> > Why?

>

> Because otherwise we could be talking about Brazillian versus German

> sausages, and who one for being placed in a very expensive bowl called

> The Cup.

Isn't the English version equally vague though?

> "typically satisfy also the predicate": swap "satisfy" and "also". That

> works.

Done.

> > The outer is adjusted accordingly, but {PA lo broda} retains the same

> > meaning.

>

> I don't see how that's possible. Before, "pa lo broda" meant *exactly*

> one broda. Now it means one *group* of broda, of indeterminate size.

> That is a massive change, unless I'm missing something.

It's {PA lo pa broda} in most contexts. Before, it meant

{PA da poi broda} exactly PA things that broda.

{pa da poi broda} could also mean exactly one group of brodas,

given the appropriate context. In most contexts, in both cases,

the usual interpretation is that we are counting individual

brodas.

> None of your examples use an outer quantifier by itself, by the way.

> Might want to fix that.

That's because in that case {lo} is elidable. There's

{mu (lo) xagji sofybakni} anyway.

> > A constant is something that always keeps the same referent. {lo

> > broda} always refers to brodas. In {mu da poi broda zo'u da brode},

> > "da" is a variable, because it takes values from the set of all things

> > that brodas. Anything with a quantifier in front takes values from the

> > set of things over which the quantifier runs.

>

> Don't tell me; tell the notes. :-)

OK.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 21:00 GMT

> > > > * "lo ctuca cu fendi lo selctu mu lo vo tadni" . the first two

> > > > lo are most definately "le" if you want it to match the English.

> > > > You should probably quantify ctuca as well.

> > >

> > > How would you translate the Lojban into normal English? (The

> > > context is a set of instructions for conducting a Lesson.)

> >

> > The way I translated it (see my other post) is "Teachers

> > (un-numbered, could be one) divide those taught into five foursomes

> > of students."

>

> I mean into normal English. The Lojban sounds normal as it is.

Teachers divide those taught into five groups of four.

> > Gender-queer-positive people would have a field day with "le", I

> > suspect.

>

> Many of them would claim they are ca'a ninmu, not that they are using

> the selbri non-veridically.

Right, and then fights would break out. That would be the field day.

:-)

> > > Is this the appropriate place and time to propose moving CMENE

> > > into BRIVLA?

> >

> > mumble, mumble It's up to you. I can't think of a *better* time,

> > though. If you do so, I will *immediately* call an extension to

> > voting, for obvious reasons.

>

> I won't then. We can always fix the definition of {la} later if that

> is ever changed.

I don't see what other BPFK section it could happen in, but we might

want to collect proposed grammar changes and do them all at once.

> > > > * lei brazo/dotco prenu, please.

> > > Why?

> >

> > Because otherwise we could be talking about Brazillian versus German

> > sausages, and who one for being placed in a very expensive bowl

> > called The Cup.

>

> Isn't the English version equally vague though?

In my dialect, "The Brazillians" can only mean "a group of Brazillian

people".

> > > The outer is adjusted accordingly, but {PA lo broda} retains the

> > > same meaning.

> >

> > I don't see how that's possible. Before, "pa lo broda" meant

> > *exactly* one broda. Now it means one *group* of broda, of

> > indeterminate size. That is a massive change, unless I'm missing

> > something.

>

> It's {PA lo pa broda} in most contexts. Before, it meant {PA da poi

> broda} exactly PA things that broda.

Please show me chapter and verse for it meaning that, please.

> {pa da poi broda} could also mean exactly one group of brodas, given

> the appropriate context. In most contexts, in both cases, the usual

> interpretation is that we are counting individual brodas.

OK, problems here. I suppose "pa lo girzu" always meant one group, but

it still means only one thing that matches the x1 of girzu. Same with

"pa da poi broda". Your proposal seems to be expanding that to one

  • group* of things that match the x1 of girzu, i.e. on group of groups.

I don't see how the old definition could ever have meant that.

> > None of your examples use an outer quantifier by itself, by the way.

> > Might want to fix that.

>

> That's because in that case {lo} is elidable. There's {mu (lo) xagji

> sofybakni} anyway.

Oh, true.

-Robni



Posted by Anonymous on Sat 05 of June, 2004 22:33 GMT

> > "typically satisfy also the predicate": swap "satisfy" and "also".

> > That works.

>

> Done.

Nope. "that satisfy the selbri typically satisfy also the predicate"

-Robin



Posted by Anonymous on Sat 05 of June, 2004 22:33 GMT


> > > "typically satisfy also the predicate": swap "satisfy" and "also".

> > > That works.

> >

> > Done.

>

> Nope. "that satisfy the selbri typically satisfy also the predicate"

?

That's what the previous version was.

Now it says:

The resulting expression indicates that the individuals or groups that satisfy

the selbri typically also satisfy the predicate for which the sumti is an

argument.

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 22:33 GMT

On Sat, Jun 05, 2004 at 02:11:25PM -0700, Jorge Llamb?as wrote:

>

> --- Robin Lee Powell wrote:

> > > > "typically satisfy also the predicate": swap "satisfy" and

> > > > "also". That works.

> > >

> > > Done.

> >

> > Nope. "that satisfy the selbri typically satisfy also the

> > predicate"

>

> ? That's what the previous version was. Now it says:

>

> The resulting expression indicates that the individuals or groups that

> satisfy the selbri typically also satisfy the predicate for which the

> sumti is an argument.

OK, that was wierd. Nevermind.

-Robin



Posted by Anonymous on Sat 05 of June, 2004 22:33 GMT

Robin:

> It seems to me that this changes *all* usage of "PA lo", because

> suddenly when you said "pa lo broda" and meant "One broda", you don't

> *get* one broda anymore, you get one *group* of broda, which is very,

> very different.

No, you get one instance of brodas. What counts as an instance

depends on context, but the most normal instances are individual

brodas, at least in cases where brodas are normally individuated.

What counts as an instance in {pa djacu} is much more context

sensitive.

But this is nothing new. {ci lo djacu} always could be three

glasses of water or three lakes, depending on context.

> su'o pa lo prenu cu prami do

> At least one person loves you.

>

> http://www.lojban.org/files/brochures/lesson4.html

>

> In your version, this means "At least one group of people loves you",

> does it not?

No, "at least one instance of people loves you". In the case

of su'o surely it makes no difference anyway, because "group"

must include groups of one.

> 19 May 2003 14:23:19 ci re'u ca pa lo cacra

>

> http://www.lojban.org/resources/irclog/lojban/2003_05_20-02_22.txt

>

> Again, your version would be "Three times in one *group* of hours".

(cire'u is the third time)

I'd say {cire'u lo cacra be li pa}.

> > > The regularity is in usage. The outer quantifier works the same for

> > > *all* gadri *except* lo.

> >

> > *all* is le and la, right?

> > It certainly doesn't work like that for loi, lei, lai, lo'i, le'i,

> > la'i, lo'e and le'e.

>

> You have very few examples of quantification of those ones. However,

> those all say "An outer quantifier can be used to indicate a subset of

> that cardinality ", or "subgroup" instead of subset. la and le say " An

> outer quantifier can be used to quantify over members of the group."

Right, so something very different.

> Ignoring lo'e and le'e, of course.

>

> If those two are different, I don't understand one or the other.

If you understand outer quantification of sets, please explain

it to me.

> The only example of quantification of these articles is "ro le verba",

> which helps very little.

>

> If "indicate a subset of that cardinality" and "quantify over members of

> the group" mean substantially different things, then on behalf of slow

> people everywhere I request more verbosity. "In other words, ..." would

> be nice. More examlpes would be nice too.

Consider the set {a, b, c}.

The subsets are:

{}, {a}, {b}, {c}, {a, b}, {b, c}, {a, c} and {a, b, c}.

Quantifying over the subsets would mean that you can say something

of up to eight objects. For example: Exactly four subsets of

{a, b, c} contain b as a member. That's NOT what quantifiers

on lo'i would seem to be for.

If the quantifier indicates a cardinality, as I wrote, then it

is not doing quantification over a set. It is not very clear at

all what it is doing, either, but presumably {pimu lo'i broda}

is something like a set with half the members of lo'i broda. There

are usually many such sets, and I don't really know how

{pimu lo'i} works. "at least one subset of half the cardinality

of the set"? Exactly one such subset? A generic such subset?

Fortunately all this is irrelevant, because we never talk about

such things in normal situations, and when we want to talk math

there are better ways of doing it (like using proper predicates

for "set", "subset", etc.)

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 22:33 GMT

Robin:

> Attempting to increase clarity.

> OK, clarifying my stance on the lo quantifier thing.

>

> It's not the inner quantifier change I'm worried about. The two or three

> times that someone has really *wanted* to quantify the entire universe of

> objects can go stuff themselves, although I *am* curious as to how to say

> "One of the pictures I can draw, of which there are only two in the universe"

> in the new lo.

lo/le rore pixra poi mi kakne lo nu finti

> What I'm worried about is the change that this effectively makes to the

> behaviour of the *outer* quantifier of lo. "An outer quantifier can be used

> to quantify over instances of the generic individual or group.": "or group"

> is what I'm worried about. "pa lo broda" used to me *exactly* one broda; now

> it can mean one *group* of broda, unless I'm missing something.

>

> This is a major and drastic change to past usage, and I cannot support it.

>

> I doubt that fixing it is difficult, but it needs to be addressed.

I changed it to: "An outer quantifier can be used to quantify over

instances of the generic individual or, if an explicit inner quantifier

is given, over instances of a group."

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 22:34 GMT

pc:

> but then how keep the quantifier from becoming part of the name — and how

> distinguish from cases where it is part of the nme "Three Cows Came> {la ci

> bakni. cu klama} Three cows or Three-Cows?

We already have that case allowed: {la ci bakni}.

That can mean "the group of three things that I call 'Cow'"

or the one thing that I call 'The Three Cows' (a restaurant, say).

I would say the second reading has to be the predominant one.

For several things with the same name it makes more sense to

use {lo ci me la bakni}.

I changed the definition of {la} accordingly to:

"An inner quantifier can be used in the case of a selbri to

indicate the cardinality of the group or as part of the name."

I didn't do the same change for lai and la'i because in those

cases an actual cardinality seems much more likely.

Anyone against?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sat 05 of June, 2004 22:34 GMT

Robin:

> > > > > * "lo ctuca cu fendi lo selctu mu lo vo tadni" .

> Teachers divide those taught into five groups of four.

Isn't "the class" a more normal translation than "those taught"?

I want some examples where English "the" is translated with

lojban {lo} so that people don't automatically assume the=le.

I don't think translating {lo ctuca} as "the teacher" in this

sentence is such a stretch. Do you think I should change:

le palta ba'o porpi i ma gasnu i xu le gerku cu go'i

The dish is broken. Who did it? Was it the dog?

to: The dishes are broken. Who did it? Was it the dogs?

> > > > Is this the appropriate place and time to propose moving CMENE

> > > > into BRIVLA?

>

> I don't see what other BPFK section it could happen in, but we might

> want to collect proposed grammar changes and do them all at once.

The special section 'Formal Grammar' would be where this belongs,

I think.

> > > > > * lei brazo/dotco prenu, please.

> > > > Why?

> > >

> > > Because otherwise we could be talking about Brazillian versus German

> > > sausages, and who one for being placed in a very expensive bowl

> > > called The Cup.

> >

> > Isn't the English version equally vague though?

>

> In my dialect, "The Brazillians" can only mean "a group of Brazillian

> people".

Ok, English is more precise in this case. That does not mean that

you can't use {lei brazo} to refer to the Brazilians.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Votes


Posted by xorxes on Sat 05 of June, 2004 22:39 GMT posts: 1912

Can we know who is voting against the proposal at this point, and what their objections are? I know pc is one, but there are two others.

ki'e mi'e xorxes



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

On Sat, Jun 05, 2004 at 03:39:27PM -0700, [email protected] wrote:

> Votes

>

> Can we know who is voting against the proposal at this point, and what

> their objections are? I know pc is one, but there are two others.

Me and Arnt, both for the lo outer quantifier over groups issue. Mine

will be reversed shortly.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

I think the point is that if what comes after {lo} is {broda} the quantifiers on {lo broda} are just quantifiers on {broda}, while if what follows is {PA broda} then quantifers on {lo pa broda} are quantifiers on groups of {PA broda}, which cannot be be written without the {lo}. In other words, with {lo} you are talking about the whole phrase that follows in the sumti and what satisfies it, while with {le} you are talking about just what is said to satisfy the brivla that follows in the sumti (and then you are told how many there are. The external quantifiers work the same in each case; it is the internal ones that play different roles.

Jorge Llambías wrote:

Robin:

> It seems to me that this changes *all* usage of "PA lo", because

> suddenly when you said "pa lo broda" and meant "One broda", you don't

> *get* one broda anymore, you get one *group* of broda, which is very,

> very different.

No, you get one instance of brodas. What counts as an instance

depends on context, but the most normal instances are individual

brodas, at least in cases where brodas are normally individuated.

What counts as an instance in {pa djacu} is much more context

sensitive.

But this is nothing new. {ci lo djacu} always could be three

glasses of water or three lakes, depending on context.

> su'o pa lo prenu cu prami do

> At least one person loves you.

>

> http://www.lojban.org/files/brochures/lesson4.html

>

> In your version, this means "At least one group of people loves you",

> does it not?

No, "at least one instance of people loves you". In the case

of su'o surely it makes no difference anyway, because "group"

must include groups of one.

> 19 May 2003 14:23:19 ci re'u ca pa lo cacra

>

> http://www.lojban.org/resources/irclog/lojban/2003_05_20-02_22.txt

>

> Again, your version would be "Three times in one *group* of hours".

(cire'u is the third time)

I'd say {cire'u lo cacra be li pa}.

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

H>This is useless unless ione has been indoctrinated into the incoherent notion of a generic individual (contradictory already) and its instances. Quantification is over brodas or groups of brodas.

Jorge Llambías wrote:

Robin:

> Attempting to increase clarity.

> OK, clarifying my stance on the lo quantifier thing.

>

> It's not the inner quantifier change I'm worried about. The two or three

> times that someone has really *wanted* to quantify the entire universe of

> objects can go stuff themselves, although I *am* curious as to how to say

> "One of the pictures I can draw, of which there are only two in the universe"

> in the new lo.

lo/le rore pixra poi mi kakne lo nu finti

> What I'm worried about is the change that this effectively makes to the

> behaviour of the *outer* quantifier of lo. "An outer quantifier can be used

> to quantify over instances of the generic individual or group.": "or group"

> is what I'm worried about. "pa lo broda" used to me *exactly* one broda; now

> it can mean one *group* of broda, unless I'm missing something.

>

> This is a major and drastic change to past usage, and I cannot support it.

>

> I doubt that fixing it is difficult, but it needs to be addressed.

H>I changed it to: "An outer quantifier can be used to quantify over

instances of the generic individual or, if an explicit inner quantifier

is given, over instances of a group."

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

I> And what distinguishes the two — very different — cases?

G> So, we can clearly use this for cmevla as well (though the use of {me} is as always questionable) and there is no problem. And notice that the "internal quantifier" now does not function like that in other sumti, which was the reason for trying to shove quantifiers in in the first place.

F>This does not solve the "problem" which you posed to begin with, and, as you note, with {lai} and {la'i} the work around for that problem will not work either.

Jorge Llambías wrote:

pc:

> but then how keep the quantifier from becoming part of the name — and how

> distinguish from cases where it is part of the nme "Three Cows Came> {la ci

> bakni. cu klama} Three cows or Three-Cows?

I>We already have that case allowed: {la ci bakni}.

That can mean "the group of three things that I call 'Cow'"

or the one thing that I call 'The Three Cows' (a restaurant, say).

G>I would say the second reading has to be the predominant one.

For several things with the same name it makes more sense to

use {lo ci me la bakni}.

I changed the definition of {la} accordingly to:

"An inner quantifier can be used in the case of a selbri to

indicate the cardinality of the group or as part of the name."

I didn't do the same change for lai and la'i because in those

cases an actual cardinality seems much more likely.

F>Anyone against?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

E> It depends (as all references do to a greater or lesser extent) on context. It is easy to imagine cases where {lo brazo} does refer to Brazilain sausages, but in contextless cases, Brazilian people are clearly the most likely (unmarked, not needed special context).

Jorge Llambías wrote:Robin:

> > > > > * lei brazo/dotco prenu, please.

> > > > Why?

> > >

> > > Because otherwise we could be talking about Brazillian versus German

> > > sausages, and who one for being placed in a very expensive bowl

> > > called The Cup.

> >

> > Isn't the English version equally vague though?

>

E> In my dialect, "The Brazillians" can only mean "a group of Brazillian

> people".

Ok, English is more precise in this case. That does not mean that

you can't use {lei brazo} to refer to the Brazilians.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 05:59 GMT

E> It depends (as all references do to a greater or lesser extent) on context. It is easy to imagine cases where {lo brazo} does refer to Brazilain sausages, but in contextless cases, Brazilian people are clearly the most likely (unmarked, not needed special context).

Jorge Llambías wrote: Robin:

> > > > > * lei brazo/dotco prenu, please.

> > > > Why?

> > >

> > > Because otherwise we could be talking about Brazillian versus German

> > > sausages, and who one for being placed in a very expensive bowl

> > > called The Cup.

> >

> > Isn't the English version equally vague though?

>

E> In my dialect, "The Brazillians" can only mean "a group of Brazillian

> people".

Ok, English is more precise in this case. That does not mean that

you can't use {lei brazo} to refer to the Brazilians.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

On Sat, Jun 05, 2004 at 03:30:45PM -0700, Jorge Llamb?as wrote:

>

> Robin:

> > > > > > * "lo ctuca cu fendi lo selctu mu lo vo tadni" .

> > Teachers divide those taught into five groups of four.

>

> Isn't "the class" a more normal translation than "those taught"?

No, that's le.

> I want some examples where English "the" is translated with lojban

> {lo} so that people don't automatically assume the=le.

Umm, "the" *does* == "le", as far as I can tell.

> I don't think translating {lo ctuca} as "the teacher" in this

> sentence is such a stretch. Do you think I should change:

>

> le palta ba'o porpi i ma gasnu i xu le gerku cu go'i

> The dish is broken. Who did it? Was it the dog?

>

> to: The dishes are broken. Who did it? Was it the dogs?

Err.

ponder

OK, I take it back. "lo selctu" == "the class" is acceptable.

> > > > > Is this the appropriate place and time to propose moving CMENE

> > > > > into BRIVLA?

> >

> > I don't see what other BPFK section it could happen in, but we might

> > want to collect proposed grammar changes and do them all at once.

>

> The special section 'Formal Grammar' would be where this belongs,

> I think.

Yep.

> > > > > > * lei brazo/dotco prenu, please.

> > > > > Why?

> > > >

> > > > Because otherwise we could be talking about Brazillian versus

> > > > German sausages, and who one for being placed in a very

> > > > expensive bowl called The Cup.

> > >

> > > Isn't the English version equally vague though?

> >

> > In my dialect, "The Brazillians" can only mean "a group of

> > Brazillian people".

>

> Ok, English is more precise in this case. That does not mean that you

> can't use {lei brazo} to refer to the Brazilians.

True. It's a matter of personal taste; my taste happens to disagree

with yours, but it's not a big deal.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

On Sat, Jun 05, 2004 at 03:07:07PM -0700, Jorge Llamb?as wrote:

>

> pc:

> > but then how keep the quantifier from becoming part of the name --

> > and how distinguish from cases where it is part of the nme "Three

> > Cows Came> {la ci bakni. cu klama} Three cows or Three-Cows?

>

> We already have that case allowed: {la ci bakni}.

>

> That can mean "the group of three things that I call 'Cow'" or the one

> thing that I call 'The Three Cows' (a restaurant, say).

It can? As far as I know, it only means the latter.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

On Sat, Jun 05, 2004 at 02:55:55PM -0700, Jorge Llamb?as wrote:

> Robin:

> > What I'm worried about is the change that this effectively makes to

> > the behaviour of the *outer* quantifier of lo. "An outer

> > quantifier can be used to quantify over instances of the generic

> > individual or group.": "or group" is what I'm worried about. "pa

> > lo broda" used to me *exactly* one broda; now it can mean one

> > *group* of broda, unless I'm missing something.

> >

> > This is a major and drastic change to past usage, and I cannot

> > support it.

> >

> > I doubt that fixing it is difficult, but it needs to be addressed.

>

> I changed it to: "An outer quantifier can be used to quantify over

> instances of the generic individual or, if an explicit inner

> quantifier is given, over instances of a group."

That'll do. Vote changed.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

On Sat, Jun 05, 2004 at 02:43:18PM -0700, Jorge Llamb?as wrote:

> Robin:

> > > > The regularity is in usage. The outer quantifier works the same

> > > > for *all* gadri *except* lo.

> > >

> > > *all* is le and la, right? It certainly doesn't work like that for

> > > loi, lei, lai, lo'i, le'i, la'i, lo'e and le'e.

> >

> > You have very few examples of quantification of those ones.

> > However, those all say "An outer quantifier can be used to indicate

> > a subset of that cardinality ", or "subgroup" instead of subset. la

> > and le say " An outer quantifier can be used to quantify over

> > members of the group."

>

> Right, so something very different.

>

> > Ignoring lo'e and le'e, of course.

> >

> > If those two are different, I don't understand one or the other.

>

> If you understand outer quantification of sets, please explain it to

> me.

>

> > The only example of quantification of these articles is "ro le

> > verba", which helps very little.

> >

> > If "indicate a subset of that cardinality" and "quantify over

> > members of the group" mean substantially different things, then on

> > behalf of slow people everywhere I request more verbosity. "In

> > other words, ..." would be nice. More examlpes would be nice too.

>

> Consider the set {a, b, c}.

>

> The subsets are:

>

> {}, {a}, {b}, {c}, {a, b}, {b, c}, {a, c} and {a, b, c}.

>

> Quantifying over the subsets would mean that you can say something

> of up to eight objects. For example: Exactly four subsets of

> {a, b, c} contain b as a member. That's NOT what quantifiers

> on lo'i would seem to be for.

Granted.

> If the quantifier indicates a cardinality, as I wrote, then it

> is not doing quantification over a set. It is not very clear at

> all what it is doing, either, but presumably {pimu lo'i broda}

> is something like a set with half the members of lo'i broda. There

> are usually many such sets, and I don't really know how

> {pimu lo'i} works. "at least one subset of half the cardinality

> of the set"? Exactly one such subset? A generic such subset?

Ouch.

> Fortunately all this is irrelevant, because we never talk about

> such things in normal situations, and when we want to talk math

> there are better ways of doing it (like using proper predicates

> for "set", "subset", etc.)

Granted. If it's a serious problem we can come back to it.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

I'm inclined to say that the inner quantifier is always

a part of the name. This would seem to agree with something

CLL says about the inner quantifier of le:

"Note that the inner quantifier of le, even when exact, need

not be truthful: le ci nanmu means what I describe as three

men, not three of what I describe as men. This follows from

the rule that what is described by a le description represents

the speaker's viewpoint rather than the objective way things are."

So the inner quantifier is part of the description in the case

of le, and part of the name in the case of la.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

On Sat, Jun 05, 2004 at 04:57:53PM -0700, Jorge Llamb?as wrote:

>

> I'm inclined to say that the inner quantifier is always a part of the

> name. This would seem to agree with something CLL says about the inner

> quantifier of le:

>

> "Note that the inner quantifier of le, even when exact, need not

> be truthful: le ci nanmu means what I describe as three men,

> not three of what I describe as men. This follows from the rule

> that what is described by a le description represents the

> speaker's viewpoint rather than the objective way things are."

>

> So the inner quantifier is part of the description in the case of le,

> and part of the name in the case of la.

Works for me. I'd rather that then being wishy-washy about it. One can

always use the lo/le PA la PA broda trick. If you're going to go that

route, though, explicitely mentioning that trick somewhere seems like a

good idea.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 06:00 GMT

Robin Lee Powell scripsit:

> > I want some examples where English "the" is translated with lojban

> > {lo} so that people don't automatically assume the=le.

>

> Umm, "the" *does* == "le", as far as I can tell.

Well, "the" gets translated "le", but "a certain" does too. "A certain cow"

means that I know the cow (+specific) but you don't (-definite).

--

Where the wombat has walked, John Cowan

it will inevitably walk again. http://www.ccil.org/~cowan



Posted by Anonymous on Sun 06 of June, 2004 16:29 GMT

A Gedankenexperiment:

I have created a thing which I call in Lojban a jvugi. There are a dozen of them and they are identical above the molecular lattice level. They are all on my table. I put some (but not all) of them into my designated box, Waldo. At this point, {lo jvugi cu nenri le tanxe o la ualdos} is true. So also is {lo jvugi naku nenri le tanxe}. But neither {lo jvugi na nenri le tanxe} or {lo jvugi nenri gi'enai nenri le tanxe} is true, the first because its negate is true, the second because it is a contradiction. We can save the latter (and, to some extent the former) by dividing the predicate to distinguish "different senses of {nenri}" one of which applies to the jvugi in the box, another to those outside. But how to do this: the jvugi are the same in all usable properties, the box is the same. I suppose we could invent some kind of deictic property which applied only to the items pointed at or the like. But this only works if we have already individuated the jvugi (by pointing)

and turns out to be aproperty whose apllication depends upon that individuation. That is, it is a complex way of separating lo jvugi into these (pointed at) and those (not). It has to be reinvented everytime we shift which jvugi are in and which out. So the properties are always isomorphic to the individuation of the jvugi. In short, the unitary lo jvugi plays not significant role in all this, only the individual jvugi. And that means that, so far as practical concerns go, {lo jvugi} behaves exactly like {su'o jvugi} — just as CLL says. And wirthout the mumbling metaphysics.



Posted by Anonymous on Sun 06 of June, 2004 19:47 GMT

All right. I'm about to do something incredibly rude.

The incredibly voluminous argumentation on this topic has been too much

for me to keep up with; I'm literally weeks behind (I actually panic

when I accidentally bring up one of the messages on my mail-reading

window, that's how scared of the topic I am). OTOH, I do want to be

sure I understand and contribute. So I am going to see if I understand

things okay and ask my one or two questions without bothering to read

all the give-and-take which may have already answered them.

I've read through the gadri proposal, and most of it makes sense. One

or two points remain, for me, though. And maybe you've already answered

them.

(1) in the proposal, a sample sentence given is {lo ctuca cu fendi lo

selctu mu lo vo tadni}/"The teacher will divide the class into five

groups of four students." This doesn't work, by my understanding of the

proposal. It should mean that "teachers in general," (something like

"all teachers"?) will divide their classes. The English translation

sounds like it's referring to some particular teacher, which would be

{le}. If I'm wrong about the meaning of the English translation, that

just means this is a bad example or a bad English rendering.

(2) is the usage of loi/lei changed? The proposal specifies that the

group satisfies the matrix predication *as a group*. This implies that

the mass does *not* inherit the properties of its elements, or at least

not all of them. Do we still say that it's true that both {loi ratcu cu

cmalu} (as rats are small) and that {loi ratcu cu barda} (as the class

is big)? The first sentence would be better expressed as {lo ratcu cu

cmalu} in the new formulation. So is it then *not* true that {loi ratcu

cu cmalu}? Does the group have to satisfy the matrix predicate jointly

but NOT singly? Or either way? Whatever the answer to this, it should

be clarified in the proposal.

(3) just some minor expenditure of grey-matter on the subject of

gadri+NU. As I read it, many (maybe most) of the prior uses of {le nu}

are probably okay, and the rest need to be {lo nu}. That's all right,

but we'll probably be using {lo nu} more than we have been in the past.

Most of the rest looks okay and makes sense to me. Hope I haven't

muddied the waters too much...

~mark



Posted by Anonymous on Sun 06 of June, 2004 21:02 GMT

~mark:

> (1) in the proposal, a sample sentence given is {lo ctuca cu fendi lo

> selctu mu lo vo tadni}/"The teacher will divide the class into five

> groups of four students." This doesn't work, by my understanding of the

> proposal. It should mean that "teachers in general," (something like

> "all teachers"?) will divide their classes. The English translation

> sounds like it's referring to some particular teacher, which would be

> {le}. If I'm wrong about the meaning of the English translation, that

> just means this is a bad example or a bad English rendering.

It's the teacher in general. It is part of a set of directions on

how to conduct a certain lesson, the speaker has no particular

teacher in mind. Perhaps it is a bad example given that so many

people have complained about it, but "the teacher" can be generic

in English.

> (2) is the usage of loi/lei changed? The proposal specifies that the

> group satisfies the matrix predication *as a group*. This implies that

> the mass does *not* inherit the properties of its elements, or at least

> not all of them. Do we still say that it's true that both {loi ratcu cu

> cmalu} (as rats are small) and that {loi ratcu cu barda} (as the class

> is big)? The first sentence would be better expressed as {lo ratcu cu

> cmalu} in the new formulation. So is it then *not* true that {loi ratcu

> cu cmalu}? Does the group have to satisfy the matrix predicate jointly

> but NOT singly? Or either way? Whatever the answer to this, it should

> be clarified in the proposal.

As I understand it, the group has to satisfy the matrix predicate

jointly, whether or not the members satisfy it. The claim is about

the group and nothing is claimed about the members individually.

I think that's what the definition says.

I suppose {loi ci ratcu cu cmalu} and {loi ki'o ratcu cu barda}, so

{loi ratcu} could be either. We need more context.

{lo ratcu cu cmalu}, "rats are small", is definitely more clear.

> (3) just some minor expenditure of grey-matter on the subject of

> gadri+NU. As I read it, many (maybe most) of the prior uses of {le nu}

> are probably okay, and the rest need to be {lo nu}. That's all right,

> but we'll probably be using {lo nu} more than we have been in the past.

Yes, and I think {lo ka} and {lo du'u} make much more sense than

with {le}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

Jorge "Llamb����������������������������������" wrote:

Wonder why my mailer keeps doing that.

>~mark:

>

>

>>(1) in the proposal, a sample sentence given is {lo ctuca cu fendi lo

>>selctu mu lo vo tadni}/"The teacher will divide the class into five

>>groups of four students." This doesn't work, by my understanding of the

>>proposal. It should mean that "teachers in general," (something like

>>"all teachers"?) will divide their classes. The English translation

>>sounds like it's referring to some particular teacher, which would be

>>{le}. If I'm wrong about the meaning of the English translation, that

>>just means this is a bad example or a bad English rendering.

>>

>>

>

>It's the teacher in general. It is part of a set of directions on

>how to conduct a certain lesson, the speaker has no particular

>teacher in mind. Perhaps it is a bad example given that so many

>people have complained about it, but "the teacher" can be generic

>in English.

>

OK, with you. So you're saying "Classes in this school will start

tomorrow morning at nine sharp, at which time the teacher [[i.e.%20each%0A%3Cbr%20/%3Eclass'%20teacher(s)|i.e. each

class' teacher(s)]] will divide the students..." I can see that. The

English is ambiguous (big surprise) and seems to be, to me, closer to

the non-example meaning than to the example meaning.

>>(2) is the usage of loi/lei changed? The proposal specifies that the

>>group satisfies the matrix predication *as a group*. This implies that

>>the mass does *not* inherit the properties of its elements, or at least

>>not all of them. Do we still say that it's true that both {loi ratcu cu

>>cmalu} (as rats are small) and that {loi ratcu cu barda} (as the class

>>is big)? The first sentence would be better expressed as {lo ratcu cu

>>cmalu} in the new formulation. So is it then *not* true that {loi ratcu

>>cu cmalu}? Does the group have to satisfy the matrix predicate jointly

>>but NOT singly? Or either way? Whatever the answer to this, it should

>>be clarified in the proposal.

>>

>>

>

>As I understand it, the group has to satisfy the matrix predicate

>jointly, whether or not the members satisfy it. The claim is about

>the group and nothing is claimed about the members individually.

>I think that's what the definition says.

>

>I suppose {loi ci ratcu cu cmalu} and {loi ki'o ratcu cu barda}, so

>{loi ratcu} could be either. We need more context.

>

OK, gotcha. That's good, and significant. But would it be true that

{loi ki'o ratcu cu cmalu}? Um... I presume not, since although the

individuals satisfy it, the group as a whole isn't small. Oog. OTOH,

{loi dembi cu nenri le patxu} is true and seems to mean exactly the same

thing as {lo dembi cu nenri le patxu}, doesn't it? (There are beans in

the pot). Except for a natlang-biased singular/plural thing I am having

trouble avoiding. Hrm. Still feels like some ambiguity or question

there, somehow.

>{lo ratcu cu cmalu}, "rats are small", is definitely more clear.

>

Right, ok, with you.

>>(3) just some minor expenditure of grey-matter on the subject of

>>gadri+NU. As I read it, many (maybe most) of the prior uses of {le nu}

>>are probably okay, and the rest need to be {lo nu}. That's all right,

>>but we'll probably be using {lo nu} more than we have been in the past.

>>

>>

>

>Yes, and I think {lo ka} and {lo du'u} make much more sense than

>with {le}.

>

Mmm... But if I'm talking about a specific picture, say, then it should

be {le ka le pixre cu melbi kei} and not {lo ka} shouldn't it? Um, maybe.

~mark



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

Mark E. Shoulson wrote:

>All right. I'm about to do something incredibly rude.

>

>The incredibly voluminous argumentation on this topic has been too much

>for me to keep up with; I'm literally weeks behind (I actually panic

>when I accidentally bring up one of the messages on my mail-reading

>window, that's how scared of the topic I am). OTOH, I do want to be

>sure I understand and contribute. So I am going to see if I understand

>things okay and ask my one or two questions without bothering to read

>all the give-and-take which may have already answered them.

>

>

I can sympathize. I've probably spent 20 hours on this topic over the

past few weeks.

>I've read through the gadri proposal, and most of it makes sense. One

>or two points remain, for me, though. And maybe you've already answered

>them.

>

>(1) in the proposal, a sample sentence given is {lo ctuca cu fendi lo

>selctu mu lo vo tadni}/"The teacher will divide the class into five

>groups of four students." This doesn't work, by my understanding of the

>proposal. It should mean that "teachers in general," (something like

>"all teachers"?) will divide their classes. The English translation

>sounds like it's referring to some particular teacher, which would be

>{le}. If I'm wrong about the meaning of the English translation, that

>just means this is a bad example or a bad English rendering.

>

>

Zu'u it's not a specific teacher, it could be any teacher. Zu'unai once

the teacher is introduced at the top of the script (which is what that

sentence comes from) subsequent references don't refer again to any

teacher, but the same one bi'unai. Zu'unainai lo is general enough to

cover le as well as other than le. Zu'unainainai assuming the speaker is

cooperative, the listener should wonder why the speaker didn't use le,

and what are they trying to signal by using lo?

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

pc:

> A Gedankenexperiment:

>

> I have created a thing which I call in Lojban a jvugi. There are a dozen of

> them and they are identical above the molecular lattice level. They are all

> on my table. I put some (but not all) of them into my designated box, Waldo.

> At this point, {lo jvugi cu nenri le tanxe o la ualdos} is true. So also

> is {lo jvugi naku nenri le tanxe}.

As long as a different spatial tense is understood, yes.

> But neither {lo jvugi na nenri le tanxe}

> or {lo jvugi nenri gi'enai nenri le tanxe} is true, the first because its

> negate is true, the second because it is a contradiction. We can save the

> latter (and, to some extent the former) by dividing the predicate to

> distinguish "different senses of {nenri}" one of which applies to the jvugi

> in the box, another to those outside.

Or the same sense of nenri with different tenses.

> But how to do this: the jvugi are the

> same in all usable properties, the box is the same. I suppose we could

> invent some kind of deictic property which applied only to the items pointed

> at or the like. But this only works if we have already individuated the

> jvugi (by pointing)

> and turns out to be aproperty whose apllication depends upon that

> individuation. That is, it is a complex way of separating lo jvugi into

> these (pointed at) and those (not). It has to be reinvented everytime we

> shift which jvugi are in and which out. So the properties are always

> isomorphic to the individuation of the jvugi. In short, the unitary lo jvugi

> plays not significant role in all this, only the individual jvugi. And that

> means that, so far as practical concerns go, {lo jvugi} behaves exactly like

> {su'o jvugi} — just as CLL says. And wirthout the mumbling metaphysics.

You can repeat the gedankenexperiment with la jvugis. This is an object

that remains unaltered in all its properties throughout all time.

At some point you put it into Waldo, and later you take it out.

So {la jvugis cu nenri la ualdos} is true, and so is {la jvugis naku

nenri la ualdos}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

On Sun, Jun 06, 2004 at 03:44:23PM -0400, Mark E. Shoulson wrote:

> All right. I'm about to do something incredibly rude.

Participation is *NEVER* rude.

> The incredibly voluminous argumentation on this topic has been too

> much for me to keep up with;

Hear hear!

> I'm literally weeks behind (I actually panic when I accidentally bring

> up one of the messages on my mail-reading window, that's how scared of

> the topic I am).

I can understand that.

> OTOH, I do want to be sure I understand and contribute. So I am going

> to see if I understand things okay and ask my one or two questions

> without bothering to read all the give-and-take which may have already

> answered them.

Don't bother; if the proposal doesn't stand on it's own, we're not done

yet.

> (1) in the proposal, a sample sentence given is {lo ctuca cu fendi lo

> selctu mu lo vo tadni}/"The teacher will divide the class into five

> groups of four students." This doesn't work, by my understanding of

> the proposal. It should mean that "teachers in general," (something

> like "all teachers"?) will divide their classes. The English

> translation sounds like it's referring to some particular teacher,

> which would be {le}. If I'm wrong about the meaning of the English

> translation, that just means this is a bad example or a bad English

> rendering.

xorxes and I did, in fact, have a substantial discussion about that.

The issue is that lo is inspecific in number, but English is *not*,

ever. "lo ctuca" refers, in general, to a non-numbered generic teacher.

If there happens to only be one around, translating it with "the" in

English isn't really a stretch.

Bear in mind that the English instructions, despite using "the", *are*

general. The person writing the instructions did not have a particular

class or teacher in mind, but he knew there would be only one of each.

My gf Dalton just pointed out that this *only* works in context.

Parent-teacher night:

"At this school, for art class the teacher divides the class into four

groups of five".

But in the vast majority of contexts the sentence given would mean

"Teachers divide students into four groups of five".

I'd like very, very much to have a note added to this example.

Something as simple as given the proper context would suffice.

Also, if you're going to use the word "will", you need "doi ctuca ko

fendi" or similar.

> (2) is the usage of loi/lei changed?

loi is, yes, WRT inner quantifiers.

I don't understand the old outer quantifiers for these, so I can't

answer that part. As far as I can tell, for example, it was not valid

to have "pa loi broda" before; only fractional outer quantifiers were

valid.

> The proposal specifies that the group satisfies the matrix predication

> *as a group*. This implies that the mass does *not* inherit the

> properties of its elements, or at least not all of them. Do we still

> say that it's true that both {loi ratcu cu cmalu} (as rats are small)

> and that {loi ratcu cu barda} (as the class is big)? The first

> sentence would be better expressed as {lo ratcu cu cmalu} in the new

> formulation. So is it then *not* true that {loi ratcu cu cmalu}?

> Does the group have to satisfy the matrix predicate jointly but NOT

> singly? Or either way? Whatever the answer to this, it should be

> clarified in the proposal.

I'm sorry, I have *never* understood these issues. :-(

> (3) just some minor expenditure of grey-matter on the subject of

> gadri+NU. As I read it, many (maybe most) of the prior uses of {le

> nu} are probably okay, and the rest need to be {lo nu}. That's all

> right, but we'll probably be using {lo nu} more than we have been in

> the past.

I already am. :-)

I don't see that the previous "le nu" cause any problems, but suddenly

we actually *understand* what "lo nu" *means*, so yeah, we'll be using

it more.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

Mark E. Shoulson wrote:

>Jorge "Llamb=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=

BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD

EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF

BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF=BD=EF=BF

=BD=EF=BF=BD=EF=BF=BD" wrote:

>

>Wonder why my mailer keeps doing that.

> =20

>

What mailer?

>OK, gotcha. That's good, and significant. But would it be true that=20

>{loi ki'o ratcu cu cmalu}? Um... I presume not, since although the=20

>individuals satisfy it, the group as a whole isn't small. Oog. OTOH,=20

>{loi dembi cu nenri le patxu} is true and seems to mean exactly the same=

=20

>thing as {lo dembi cu nenri le patxu}, doesn't it? (There are beans in=20

>the pot). Except for a natlang-biased singular/plural thing I am having=

=20

>trouble avoiding. Hrm. Still feels like some ambiguity or question=20

>there, somehow.

> =20

>

This example sneakily introduces the idea of Substance because as a=20

food, "beans" sounds like "beef" or "water", those innately continuous=20

materials for which loi has been used (the third sense of loi!!) but for=20

which we now tend to use tu'o as a quantifier {tu'o dembi, tu'o bakni}=20

signalling that counting chunks is meaningless. Beans are case where the=20

item is actually countable, and unlike cows, are actually consumed in=20

their countable state (not requiring butchery) but usually not counted=20

as they are being used, hence are treated as a substance measured by=20

mass or volume.

>>>(3) just some minor expenditure of grey-matter on the subject of=20

>>>gadri+NU. As I read it, many (maybe most) of the prior uses of {le nu=

}=20

>>>are probably okay, and the rest need to be {lo nu}. That's all right,=

=20

>>>but we'll probably be using {lo nu} more than we have been in the past=

..

>>> =20

>>>

>>> =20

>>>

>>Yes, and I think {lo ka} and {lo du'u} make much more sense than=20

>>with {le}.

>>

>> =20

>>

>Mmm... But if I'm talking about a specific picture, say, then it should=20

>be {le ka le pixre cu melbi kei} and not {lo ka} shouldn't it? Um, maybe=

..

> =20

>

I have used lo ka to mean something-ness, and le ka to hint that I have=20

a specific property in mind. lo kamckafi is coffee-ness, but le kamckafi=20

is caffeine!

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

John E Clifford scripsit:

> In short, the unitary lo jvugi plays not significant role in all this,

> only the individual jvugi. And that means that, so far as practical

> concerns go, {lo jvugi} behaves exactly like {su'o jvugi} — just as

> CLL says. And wirthout the mumbling metaphysics.

Nobody denies that if su'o jvugi applies, lo jvugi does also. In fact,

I think there is now a hierarchy of applicability:

lo > le (except in non-veridical uses of le)

lo > loi

le > lei

BTW, as currently defined, lo'i ratcu means the set of all rats rather than

a set of some rats. I think the latter is more consistent with the

definitions for loi and lo.

--

Híggledy-pìggledy / XML programmers John Cowan

Try to escape those / I-eighteen-N woes; http://www.ccil.org/~cowan

Incontrovertibly / What we need more of is http://www.reutershealth.com

Unicode weenies and / François Yergeaus. [email protected]



Posted by Anonymous on Sun 06 of June, 2004 21:57 GMT

~mark:

>

> >It's the teacher in general. It is part of a set of directions on

> >how to conduct a certain lesson, the speaker has no particular

> >teacher in mind. Perhaps it is a bad example given that so many

> >people have complained about it, but "the teacher" can be generic

> >in English.

> >

> OK, with you. So you're saying "Classes in this school will start

> tomorrow morning at nine sharp, at which time the teacher [[i.e.%20each%0A%3Cbr%20/%3E%3E%20class'%20teacher(s)|i.e. each

> class' teacher(s)]] will divide the students..." I can see that. The

> English is ambiguous (big surprise) and seems to be, to me, closer to

> the non-example meaning than to the example meaning.

Yes. It's actually even more general than that:

"LESSON SUMMARY: 30 minutes

The teacher will ask the students to brainstorm ideas for each column.

For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim…)

The teacher will record an answer on each piece of oak tag in the column.

The teacher will divide the class into five groups of four students.

..."

From: http://www.teachers.net/lessons/posts/2450.html(external link)

So it's not even about a specific group of teachers tomorrow morning

at nine. Out of context the more specific meaning seems to be more

salient.

How about if I change it to:

cimai lo ctuca cu fendi lo selctu mu lo vo tadni

"Step 3: The teacher will divide the class

into five groups of four students."

> >I suppose {loi ci ratcu cu cmalu} and {loi ki'o ratcu cu barda}, so

> >{loi ratcu} could be either. We need more context.

> >

> OK, gotcha. That's good, and significant. But would it be true that

> {loi ki'o ratcu cu cmalu}? Um... I presume not, since although the

> individuals satisfy it, the group as a whole isn't small. Oog. OTOH,

> {loi dembi cu nenri le patxu} is true and seems to mean exactly the same

> thing as {lo dembi cu nenri le patxu}, doesn't it? (There are beans in

> the pot). Except for a natlang-biased singular/plural thing I am having

> trouble avoiding. Hrm. Still feels like some ambiguity or question

> there, somehow.

Yes, there isn't really any significant distinction in the proposal

between {lo} and {loi} for that case. The only difference I can detect

for {loi} is in how the outer quantifier behaves, as something that

is not a quantifier at all. That's why I say that lo/le/la are really

all the gadri we need.

> >Yes, and I think {lo ka} and {lo du'u} make much more sense than

> >with {le}.

> >

> Mmm... But if I'm talking about a specific picture, say, then it should

> be {le ka le pixre cu melbi kei} and not {lo ka} shouldn't it? Um, maybe.

I don't think you can have {ka} without an open slot for the thing

with the property. {lo ka le pixra cu melbi} would be the property

of consideryng the picture beautiful. For example:

do zmadu mi lo ka le pixra cu melbi (ce'u)

You more than I consider the picture beautiful.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 22:20 GMT

Jorge Llamb=EDas wrote:

>

>Yes. It's actually even more general than that:

>"LESSON SUMMARY: 30 minutes

>

>The teacher will ask the students to brainstorm ideas for each column.=20

>For example; characters- Scooby Doo, Mr. Smith, Ryan, Uncle Tim=85)

>

>The teacher will record an answer on each piece of oak tag in the column=

..

>

>The teacher will divide the class into five groups of four students.

>

>..."

>From: http://www.teachers.net/lessons/posts/2450.html(external link)

>

>So it's not even about a specific group of teachers tomorrow morning=20

>at nine. Out of context the more specific meaning seems to be more

>salient.=20

>

>How about if I change it to:

>

> cimai lo ctuca cu fendi lo selctu mu lo vo tadni

> "Step 3: The teacher will divide the class

> into five groups of four students."

> =20

>

This makes it even clearer that you are referring to a specific (albeit=20

unnamed and hypothetical) teacher and his specific classroom.

>>>Yes, and I think {lo ka} and {lo du'u} make much more sense than=20

>>>with {le}.

>>>

>>> =20

>>>

>>Mmm... But if I'm talking about a specific picture, say, then it should=

=20

>>be {le ka le pixre cu melbi kei} and not {lo ka} shouldn't it? Um, mayb=

e.

>> =20

>>

>

>I don't think you can have {ka} without an open slot for the thing

>with the property. {lo ka le pixra cu melbi} would be the property

>of consideryng the picture beautiful. For example:

>

> do zmadu mi lo ka le pixra cu melbi (ce'u)

> You more than I consider the picture beautiful.

> =20

>

I had assumed he wanted to put the ce'u in melbi3, since that requires a =

ka.

ka'e frili casnu lo ka le pixra cu melbi fi ce'u

It's easy to express what's so lovely about that picture.

--=20

Motorists honked in celebration in this Ramadi as news spread of the assa=

ssination of the president of the Iraqi Governing Council Ezzidin Salim M=

onday. "The GC is nothing," one man shouted. "They are not the Governing =

Council. They are the Prostitution Council."=20



Posted by Anonymous on Sun 06 of June, 2004 22:20 GMT

A> Spatial tenses be damned! These are true in a thoroughly untenseeed way. Nor is there anyway to rectify matters with spatial tenses other than to say "where the in-box jvugi are" and "where the out-box jvugi are".

B> See above, it always comes back to the different jvugi.

C> For la jvugis we have temporal properties that do not depend upon where the critter is.

To expand a bit on the Ge: from time to time I scoop some jvugi out of the box or toss some in or both. At any given time, some are in and some are not — not necesarily the same numbers in each case and certainly not the same individuals (not that I can really check that). All of the claims below remain true withwhatever tense you want to put in that apply (sometimes, right now, usuually, always,...) . How is this different from the situation with Mr. Jvugi?

Jorge Llambías wrote:

pc:

> A Gedankenexperiment:

>

> I have created a thing which I call in Lojban a jvugi. There are a dozen of

> them and they are identical above the molecular lattice level. They are all

> on my table. I put some (but not all) of them into my designated box, Waldo.

> At this point, {lo jvugi cu nenri le tanxe o la ualdos} is true. So also

> is {lo jvugi naku nenri le tanxe}.

A>As long as a different spatial tense is understood, yes.

> But neither {lo jvugi na nenri le tanxe}

> or {lo jvugi nenri gi'enai nenri le tanxe} is true, the first because its

> negate is true, the second because it is a contradiction. We can save the

> latter (and, to some extent the former) by dividing the predicate to

> distinguish "different senses of {nenri}" one of which applies to the jvugi

> in the box, another to those outside.

B>Or the same sense of nenri with different tenses.

> But how to do this: the jvugi are the

> same in all usable properties, the box is the same. I suppose we could

> invent some kind of deictic property which applied only to the items pointed

> at or the like. But this only works if we have already individuated the

> jvugi (by pointing)

> and turns out to be aproperty whose apllication depends upon that

> individuation. That is, it is a complex way of separating lo jvugi into

> these (pointed at) and those (not). It has to be reinvented everytime we

> shift which jvugi are in and which out. So the properties are always

> isomorphic to the individuation of the jvugi. In short, the unitary lo jvugi

> plays not significant role in all this, only the individual jvugi. And that

> means that, so far as practical concerns go, {lo jvugi} behaves exactly like

> {su'o jvugi} — just as CLL says. And wirthout the mumbling metaphysics.

C>You can repeat the gedankenexperiment with la jvugis. This is an object

that remains unaltered in all its properties throughout all time.

At some point you put it into Waldo, and later you take it out.

So {la jvugis cu nenri la ualdos} is true, and so is {la jvugis naku

nenri la ualdos}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 22:20 GMT

My point is that the opposit is true as well: if {lo jvugi} applies, so does {su'o jvugi}.

I gather that the last two implication you list are also in didpute.

John Cowan wrote:

..

Nobody denies that if su'o jvugi applies, lo jvugi does also. In fact,

I think there is now a hierarchy of applicability:

lo > le (except in non-veridical uses of le)

lo > loi

le > lei



Posted by Anonymous on Sun 06 of June, 2004 22:20 GMT

John E Clifford scripsit:

> My point is that the opposit is true as well: if {lo jvugi} applies, so does {su'o jvugi}.

Almost always, yes. I can't figure out the claxu examples, and I personally

wouldn't say lo -unicorn, but I don't think it's actually *confusing*.

As long as lo broda can't include no broda, then I think the new

definition of lo is unproblematic.

--

There is / One art John Cowan

No more / No less http://www.reutershealth.com

To do / All things http://www.ccil.org/~cowan

With art- / Lessness — Piet Hein



Posted by Anonymous on Sun 06 of June, 2004 22:20 GMT

pc:

> For la jvugis we have temporal properties that do not depend upon where

> the critter is.

Such as?

> To expand a bit on the Ge: from time to time I scoop some jvugi out of the

> box or toss some in or both. At any given time, some are in and some are

> not — not necesarily the same numbers in each case and certainly not the

> same individuals (not that I can really check that). All of the claims below

> remain true withwhatever tense you want to put in that apply (sometimes,

> right now, usuually, always,...) . How is this different from the situation

> with Mr. Jvugi?

It is the same, as far as I can tell. Here there are jvugis inside

of the box, and over there there are jvugis that are not inside of

the box.

I don't really see any insurmountable distinction between lo jvugi

and its identical instances in different places, and la jvugis and

its identical stages in different times.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 23:31 GMT

John Cowan:

> BTW, as currently defined, lo'i ratcu means the set of all rats rather than

> a set of some rats. I think the latter is more consistent with the

> definitions for loi and lo.

I tried to keep the definitions of loi/lei/lai/lo'i/le'i/la'i/lo'e/le'e

as close to the old ones as I could, but I wouldn't mind making that

change, unless people are opposed.

One problem I see is that "the set of all rats" is pretty well defined,

whereas "a set of some rats" could mean:

1 - a generic set of rats.

2 - at least one set of rats.

3 - exactly one set of rats.

I would discard the third option, and if the outer quantifier

were used to quantify over sets of rats, then we can let the

unquantified one be a generic set. {ci lo'i mu ratcu} would then

be "three sets of 5 rats each". (The sets might have non-empty

intersections.)

{lo'i ro ratcu} is still available for the set of all rats so

no expressive power is lost.

Comments?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 23:31 GMT

Hey, if it can't include {no broda} then it must include {su'o broda}, those are the choices. And then, of course, it is not a new definition but just the old one in weird clothes.

John Cowan wrote:John E Clifford scripsit:

> My point is that the opposit is true as well: if {lo jvugi} applies, so does {su'o jvugi}.

Almost always, yes. I can't figure out the claxu examples, and I personally

wouldn't say lo -unicorn, but I don't think it's actually *confusing*.

As long as lo broda can't include no broda, then I think the new

definition of lo is unproblematic.

--

There is / One art John Cowan

No more / No less http://www.reutershealth.com

To do / All things http://www.ccil.org/~cowan

With art- / Lessness — Piet Hein



Posted by Anonymous on Sun 06 of June, 2004 23:31 GMT

John E Clifford scripsit:

> Hey, if it can't include {no broda} then it must include {su'o broda}, those are the choices. And then, of course, it is not a new definition but just the old one in weird clothes.

As may be. My job is not to make the best possible definition, but

to decide if I can live with the new one.

--

John Cowan [email protected] www.ccil.org/~cowan www.reutershealth.com

And now here I was, in a country where a right to say how the country should

be governed was restricted to six persons in each thousand of its population.

For the nine hundred and ninety-four to express dissatisfaction with the

regnant system and propose to change it, would have made the whole six

shudder as one man, it would have been so disloyal, so dishonorable, such

putrid black treason. --Mark Twain's Connecticut Yankee



Posted by Anonymous on Sun 06 of June, 2004 23:31 GMT

A> Now, yesterday, tonorrow, etc. none of which make reference to la jvugis.

B> Over where? Can you specify the location independently of its being the place where the out-of-box jvugi are?

C>I despair of ever gettting you to admit what is so plain and obvious, but to say it again (without the plain and obvious part about la jvugis being given and its slices being created and the instances of jvugi beiung given and Mr. Jvugi being created), in the case of contradictory properties, I can disambiguate the predicate on {la jvugis} without reference to what la jvugis is doeing, etc (different times of assignable sorts) but I cannot disambiguate the predicate on {lo jvugis} without referring — precisely or vaguely — to the particular instances of lo jvugis involved. It is only the instances that mark out places s significant, the times are there and separated out already.

Jorge Llambías wrote:

pc:

> For la jvugis we have temporal properties that do not depend upon where

> the critter is.

A>Such as?

> To expand a bit on the Ge: from time to time I scoop some jvugi out of the

> box or toss some in or both. At any given time, some are in and some are

> not — not necesarily the same numbers in each case and certainly not the

> same individuals (not that I can really check that). All of the claims below

> remain true withwhatever tense you want to put in that apply (sometimes,

> right now, usuually, always,...) . How is this different from the situation

> with Mr. Jvugi?

B>It is the same, as far as I can tell. Here there are jvugis inside

of the box, and over there there are jvugis that are not inside of

the box.

C>I don't really see any insurmountable distinction between lo jvugi

and its identical instances in different places, and la jvugis and

its identical stages in different times.

mu'o mi'e xorxes



Posted by Anonymous on Sun 06 of June, 2004 23:31 GMT

Jorge Llambías wrote:

John Cowan:

> BTW, as currently defined, lo'i ratcu means the set of all rats rather than

> a set of some rats. I think the latter is more consistent with the

> definitions for loi and lo.

I tried to keep the definitions of loi/lei/lai/lo'i/le'i/la'i/lo'e/le'e

as close to the old ones as I could, but I wouldn't mind making that

change, unless people are opposed.

One problem I see is that "the set of all rats" is pretty well defined,

whereas "a set of some rats" could mean:

1 - a generic set of rats.

2 - at least one set of rats.

3 - exactly one set of rats.

I would discard the third option, and if the outer quantifier

were used to quantify over sets of rats, then we can let the

unquantified one be a generic set. {ci lo'i mu ratcu} would then

be "three sets of 5 rats each". (The sets might have non-empty

intersections.)

{lo'i ro ratcu} is still available for the set of all rats so

no expressive power is lost.

Comments?

mu'o mi'e xorxes

Aside from not being clear about what is the difference between 1 and 2, this seems sensible and coherent — if we have shifted the meaning of the internal quantifiers



Posted by Anonymous on Sun 06 of June, 2004 23:54 GMT

On Sun, Jun 06, 2004 at 03:38:03PM -0700, Jorge Llamb?as wrote:

lo'i usage

> I would discard the third option, and if the outer quantifier were

> used to quantify over sets of rats, then we can let the unquantified

> one be a generic set.

Thus matching lo.

> {ci lo'i mu ratcu} would then be "three sets of 5 rats each". (The

> sets might have non-empty intersections.)

>

> {lo'i ro ratcu} is still available for the set of all rats so no

> expressive power is lost.

>

> Comments?

Looks good to me, but I think I'll have to call another extension. No

great loss, though.

Certainly having the old "lo'i mu ratcu" means a set of the exactly five

rats in the world seems amazingly pointless.

-Robin



Posted by Anonymous on Sun 06 of June, 2004 23:54 GMT

Robin:

> Looks good to me, but I think I'll have to call another extension. No

> great loss, though.

>

> Certainly having the old "lo'i mu ratcu" means a set of the exactly five

> rats in the world seems amazingly pointless.

And if you ever need to say it, you still can: {lo'i romu ratcu}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 23:54 GMT

pc:

> A> Now, yesterday, tonorrow, etc. none of which make reference to la jvugis.

Here, there, over there, they make no reference to lo jvugis either.

> B> Over where? Can you specify the location independently of its being the

> place where the out-of-box jvugi are?

If you can specify the time independently of its being the time

when la jvugis is out of the box, then I can specify the location, yes.

> , in the case of contradictory properties, I can

> disambiguate the predicate on {la jvugis} without reference to what la jvugis

> is doeing, etc (different times of assignable sorts) but I cannot

> disambiguate the predicate on {lo jvugis} without referring — precisely or

> vaguely — to the particular instances of lo jvugis involved. It is only the

> instances that mark out places s significant, the times are there and

> separated out already.

I don't see how times are more significantly separated than places.

You have times when la jvugis is out of the box, and times when it

is inside.

You have places where lo jvugis is out of the box, and places where it

is inside.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Sun 06 of June, 2004 23:54 GMT

On Sun, Jun 06, 2004 at 06:04:25PM -0400, xod wrote:

> Jorge Llamb=EDas wrote:

snip

> >

> >The teacher will divide the class into five groups of four students.

> >

> >..."

> >

> >So it's not even about a specific group of teachers tomorrow morning

> >at nine. Out of context the more specific meaning seems to be more

> >salient.

> >

> >How about if I change it to:

> >

> > cimai lo ctuca cu fendi lo selctu mu lo vo tadni "Step 3: The

> > teacher will divide the class into five groups of four students."

>

>

> This makes it even clearer that you are referring to a specific

> (albeit unnamed and hypothetical) teacher and his specific classroom.

Is that an objection, xod? If so, I don't understand what you are

suggesting.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 00:09 GMT

A> There all deictic and, if they are going to do the work wanted, what is pointed at has to be various instnaces of lo jvugi. Otherwise, {lo jvugi here, etc. naku nenri le tanxe} will generally be false, while {lo jvugi naku nenri le tanxe} is true, yet is said to be explained by the other. And, of course, when we take the case over a period of time, none of the locations will work.

B> I am given a time and I look to see what is the situation with la jvugis. But the places that I need to look at are not given before hand. I pick a place and there are no jvugi there this time, so it does not explain {lo jvugi naku nenri le tanxe} or the place turns out to be inside the box this time. Time and space do not behave in exactly the same way.

C> The issue is separating the two predicates by spatial tenses in such a way as to make it possible to go from {lo jvugi X nenri le tanxe ije lo jvugi Ynaku nenri le tanxe} to {lo jvugi X nenri gi'e (or whatever it takes) Y naku nenri le tanxe}, without reference, over or covert, to where instances of lo jvugi are.

pc:

> A> Now, yesterday, tonorrow, etc. none of which make reference to la jvugis.

A>Here, there, over there, they make no reference to lo jvugis either.

> B> Over where? Can you specify the location independently of its being the

> place where the out-of-box jvugi are?

B>If you can specify the time independently of its being the time

when la jvugis is out of the box, then I can specify the location, yes.

> , in the case of contradictory properties, I can

> disambiguate the predicate on {la jvugis} without reference to what la jvugis

> is doeing, etc (different times of assignable sorts) but I cannot

> disambiguate the predicate on {lo jvugis} without referring — precisely or

> vaguely — to the particular instances of lo jvugis involved. It is only the

> instances that mark out places s significant, the times are there and

> separated out already.

C>don't see how times are more significantly separated than places.

You have times when la jvugis is out of the box, and times when it

is inside.

You have places where lo jvugis is out of the box, and places where it

is inside.

mu'o mi'e xorxes



Posted by Anonymous on Mon 07 of June, 2004 00:27 GMT

Robin Lee Powell wrote:

>On Sun, Jun 06, 2004 at 06:04:25PM -0400, xod wrote:

>

>

>>Jorge Llamb=EDas wrote:

>>

>>

>snip

>

>

>>>The teacher will divide the class into five groups of four students.

>>>

>>>..."

>>>

>>>So it's not even about a specific group of teachers tomorrow morning

>>>at nine. Out of context the more specific meaning seems to be more

>>>salient.

>>>

>>>How about if I change it to:

>>>

>>> cimai lo ctuca cu fendi lo selctu mu lo vo tadni "Step 3: The

>>> teacher will divide the class into five groups of four students."

>>>

>>>

>>This makes it even clearer that you are referring to a specific

>>(albeit unnamed and hypothetical) teacher and his specific classroom.

>>

>>

>

>Is that an objection, xod? If so, I don't understand what you are

>suggesting.

>

>

It's a (minor) objection to the wording of that example, but not to the

proposal.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."


Voting


Posted by noras on Mon 07 of June, 2004 00:27 GMT posts: 23

I am voting "no" at this point. This does not mean I am unalterably opposed. The volume of traffic has been too much for me to keep up with to be able to make a real decision.

I think that the original 2-week time-period was unrealistic, especially for this controversial a topic. In addition, after any changes or significant volume of discussion, there should be at least 2 weeks to re-review everthing; this will also give some leeway for people, like me, who don't spend all day every day reading e-mail or the tiki. There have been 28 messages since Robin posted the latest extension.



Posted by Anonymous on Mon 07 of June, 2004 00:49 GMT

pc:

> I am given a time and I look to see what is the situation with la jvugis.

> But the places that I need to look at are not given before hand. I pick a

> place and there are no jvugi there this time, so it does not explain {lo

> jvugi naku nenri le tanxe} or the place turns out to be inside the box this

> time. Time and space do not behave in exactly the same way.

Let's say la jvugis has a limited duration then, so that there are times

when it does not exist, just as there are places where there are no jvugi.

> C> The issue is separating the two predicates by spatial tenses in such a

> way as to make it possible to go from {lo jvugi X nenri le tanxe ije lo jvugi

> Ynaku nenri le tanxe} to {lo jvugi X nenri gi'e (or whatever it takes) Y naku

> nenri le tanxe}, without reference, over or covert, to where instances of lo

> jvugi are.

How do you make the equivalent claim for la jvugis without reference,

overt or covert, to times when there are stages of la jvugis?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 01:35 GMT

xod scripsit:

> This example sneakily introduces the idea of Substance because as a

> food, "beans" sounds like "beef" or "water", those innately continuous

> materials for which loi has been used (the third sense of loi!!) but for

> which we now tend to use tu'o as a quantifier {tu'o dembi, tu'o bakni}

> signalling that counting chunks is meaningless. Beans are case where the

> item is actually countable, and unlike cows, are actually consumed in

> their countable state (not requiring butchery) but usually not counted

> as they are being used, hence are treated as a substance measured by

> mass or volume.

IMHO using loi for substances is not a third use at all, but the same as

using it for aggregates: the countability of the individual components

is irrelevant, and in fact may constitute a slippery slope that we

should avoid trying to put a fence in.

A substance can be seen as made up of component parts, and this can

even be true in a physical sense: the water in a puddle of rain got

there drop by drop. If you look from far enough away, you can't tell

whether the "sailor all over the deck" is literally goo, or just

an aggregation of sailors standing still.

I would ask, therefore, that the section on substances be removed

from the proposal, because I think it is strictly irrelevant and

changes existing precedent. I don't object to referring to water-the-

substance as "lo djacu", but I don't think it necessary to rule out

the use of "loi" here even by implication.

--

Is a chair finely made tragic or comic? Is the John Cowan

portrait of Mona Lisa good if I desire to see [email protected]

it? Is the bust of Sir Philip Crampton lyrical, www.ccil.org/~cowan

epical or dramatic? If a man hacking in fury www.reutershealth.com

at a block of wood make there an image of a cow,

is that image a work of art? If not, why not? --Stephen Dedalus



Posted by Anonymous on Mon 07 of June, 2004 01:35 GMT

John Cowan:

> A substance can also be seen as made up of component parts, and this can

> even be true in a physical sense: the water in a puddle of rain got

> there drop by drop. If you look from far enough away, you can't tell

> whether the "sailor all over the deck" is literally goo, or just

> an aggregation of sailors standing still.

I added that paragraph to the notes. The "also" is mine.

> I would ask, therefore, that the section on substances be removed

> from the proposal, because I think it is strictly irrelevant and

> changes existing precedent. I don't object to referring to water-the-

> substance as "lo djacu", but I don't think it necessary to rule out

> the use of "loi" here even by implication.

What is the status of the "Notes" section? I understand it as

not being a part of the definitions we're voting on, but just

for comments and clarifications. Robin?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 02:34 GMT

Jorge Llamb?as scripsit:

> What is the status of the "Notes" section? I understand it as

> not being a part of the definitions we're voting on, but just

> for comments and clarifications. Robin?

In that case, I'd like this to be added to the definitions of loi and lei:

(A substance may be treated as a group made up of

individuals without worrying about which individuals

they are.)

This should appear just before the text about quantifiers.

--

John Cowan [email protected] www.ccil.org/~cowan www.reutershealth.com

I must confess that I have very little notion of what [[s.%204%20of%20the%20British%0A%3Cbr%20/%3ETrade%20Marks%20Act,%201938|s. 4 of the British

Trade Marks Act, 1938]] is intended to convey, and particularly the sentence

of 253 words, as I make them, which constitutes sub-section 1. I doubt if

the entire statute book could be successfully searched for a sentence of

equal length which is of more fuliginous obscurity. --MacKinnon LJ, 1940



Posted by Anonymous on Mon 07 of June, 2004 02:34 GMT

John Cowan wrote:

>xod scripsit:

>

>

>

>>This example sneakily introduces the idea of Substance because as a

>>food, "beans" sounds like "beef" or "water", those innately continuous

>>materials for which loi has been used (the third sense of loi!!) but for

>>which we now tend to use tu'o as a quantifier {tu'o dembi, tu'o bakni}

>>signalling that counting chunks is meaningless. Beans are case where the

>>item is actually countable, and unlike cows, are actually consumed in

>>their countable state (not requiring butchery) but usually not counted

>>as they are being used, hence are treated as a substance measured by

>>mass or volume.

>>

>>

>

>IMHO using loi for substances is not a third use at all, but the same as

>using it for aggregates: the countability of the individual components

>is irrelevant, and in fact may constitute a slippery slope that we

>should avoid trying to put a fence in.

>

>A substance can be seen as made up of component parts, and this can

>even be true in a physical sense: the water in a puddle of rain got

>there drop by drop. If you look from far enough away, you can't tell

>whether the "sailor all over the deck" is literally goo, or just

>an aggregation of sailors standing still.

>

>

Sailors can act as a collective, but the number of sailors involved is

often very important. "One sailor" is a meaningful idea. The number of

sailors can be asked and answered.

One drop of water plus one drop of water equals one drop. Countability

here is not irrelevant, it's absolutely meaningless! "pa djacu" doesn't

mean a single water molecule, so while tu'o blosoi emerges from so'i

blosoi, tu'o djacu does not emerge from so'i djacu.

And the invocation of molecules could be considered a bias towards an

atomic metaphysics, justified only by very recent developments in science.

There are cases where many individuals could appear as and be treated as

a substance, but there are many substances for which decomposition is

problematic, therefore the concept of substance remains distinct from

emergent properties or pluralities. The continued conflation should be

discouraged.

>I would ask, therefore, that the section on substances be removed

>from the proposal, because I think it is strictly irrelevant and

>changes existing precedent.

>

"Precedent" is the embarrassing fusion of plurality, collective, and

substance. The proposal clarifies this in the least destructive manner

possible.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Mon 07 of June, 2004 02:34 GMT

xod scripsit:

> "Precedent" is the embarrassing fusion of plurality, collective, and

> substance. The proposal clarifies this in the least destructive manner

> possible.

I think consolidating substance with (I don't know the term

for it, the three guys who carry the piano) is entirely correct, becaue

no fixed distinction can be maintained.

--

John Cowan [email protected] www.ccil.org/~cowan www.reutershealth.com

The penguin geeks is happy / As under the waves they lark

The closed-source geeks ain't happy / They sad cause they in the dark

But geeks in the dark is lucky / They in for a worser treat

One day when the Borg go belly-up / Guess who wind up on the street.



Posted by Anonymous on Mon 07 of June, 2004 02:34 GMT

John Cowan wrote:

>xod scripsit:

>

>

>

>>"Precedent" is the embarrassing fusion of plurality, collective, and

>>substance. The proposal clarifies this in the least destructive manner

>>possible.

>>

>>

>

>I think consolidating substance with (I don't know the term

>for it, the three guys who carry the piano) is entirely correct, becaue

>no fixed distinction can be maintained.

>

>

>

You've deleted and ignored all my arguments. Am I expected to re-paste

them, or concede my point to your superior skills on that fierce delete key?

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Mon 07 of June, 2004 02:34 GMT

xod scripsit:

> You've deleted and ignored all my arguments. Am I expected to re-paste

> them, or concede my point to your superior skills on that fierce delete key?

No. I'm just telling you what I think, not trying to convince you otherwise.

(In general, this is one of the most uncommunicable facts via the Internet;

it's hard enough to communicate it in person.)

--

"In my last lifetime, John Cowan

I believed in reincarnation; http://www.ccil.org/~cowan

in this lifetime, [email protected]

I don't." --Thiagi http://www.reutershealth.com



Posted by Anonymous on Mon 07 of June, 2004 12:16 GMT

On Sun, Jun 06, 2004 at 06:26:14PM -0700, Jorge Llamb?as wrote:

> > I would ask, therefore, that the section on substances be removed

> > from the proposal, because I think it is strictly irrelevant and

> > changes existing precedent. I don't object to referring to

> > water-the- substance as "lo djacu", but I don't think it necessary

> > to rule out the use of "loi" here even by implication.

>

> What is the status of the "Notes" section? I understand it as not

> being a part of the definitions we're voting on, but just for comments

> and clarifications. Robin?

The Notes are very much a part of what we are voting on. We are voting

on the whole page. Having said that, they are not a part of the

definitions themselves. They have, in my mind, about the same value as

a piece of officially produced teaching material, or the "usage

convention" notes in the Red Book.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 12:16 GMT

On Sun, Jun 06, 2004 at 05:27:36PM -0700, [email protected] wrote:

> I am voting "no" at this point. This does not mean I am unalterably

> opposed. The volume of traffic has been too much for me to keep up

> with to be able to make a real decision.

I will be responding more fully to this when I can do so with some

modicum of politeness, but one point needs to be made immediatly:

The discussion is irrelevant. Utterly, totally, completely irrelevant.

At least for purposes of casting a vote.

The proposal *must* stand on its own, without the discussion. Read the

proposal (it's not long), and go ahead and ask your questions, if any.

I, for one, don't care in the slightest if you ask questions that have

been asked before. Heck, I'll even call you on my dime to answer your

questions if you like.

The important thing is, though, that the proposal must stand on its own,

or we aren't done yet. Save your sanity, ignore the discussion.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 13:27 GMT

I apologize for getting sidetracked on xorxes' red herring here. The point is that {lo jvugi} (and {la jvugis} for that matter) is not transparent to conjunction. The need to go herring off to find another argument that seems not to be problematic in this way merely emphasizes the basic point: {lo jvugi} in any sense is not transparent to logical operations (we have not looked at quantifiers yet but the same non-transparency applies — that is, as with "and." it works in the one direction it does for "some" but not in the other).

Nor need it be. No language I know of (all I-E alas) has anything that works in the way this {lo broda} construction is meant to. And none needs one, getting by just fine with a variety of other devices, most of which Lojban already has, either in place or needed for other purposes anyhow. In addition to which, the proposed {lo broda} does not even solve the two "problems" (i.e., trivial rules that people are either too ignorant or to lazy to apply) it claims to solve: it does not eliminate or otherwise deal with intensional contexts and it does not allow free movement across quantifier and negation boundaries. I say leave {lo} as it is and learn to say what you mean.

Jorge Llambías wrote:

pc:

> I am given a time and I look to see what is the situation with la jvugis.

> But the places that I need to look at are not given before hand. I pick a

> place and there are no jvugi there this time, so it does not explain {lo

> jvugi naku nenri le tanxe} or the place turns out to be inside the box this

> time. Time and space do not behave in exactly the same way.

Let's say la jvugis has a limited duration then, so that there are times

when it does not exist, just as there are places where there are no jvugi.

> C> The issue is separating the two predicates by spatial tenses in such a

> way as to make it possible to go from {lo jvugi X nenri le tanxe ije lo jvugi

> Ynaku nenri le tanxe} to {lo jvugi X nenri gi'e (or whatever it takes) Y naku

> nenri le tanxe}, without reference, over or covert, to where instances of lo

> jvugi are.

How do you make the equivalent claim for la jvugis without reference,

overt or covert, to times when there are stages of la jvugis?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 13:59 GMT

Robin:

> The Notes are very much a part of what we are voting on. We are voting

> on the whole page. Having said that, they are not a part of the

> definitions themselves. They have, in my mind, about the same value as

> a piece of officially produced teaching material, or the "usage

> convention" notes in the Red Book.

Given that, I think the notes is the right place to present the different

ways of viewing substances. I don't see any conflicts arising from

seeing it one or the other. Let's say we use {lo broda} in a context

that calls for a substance and someone asks {lo xo broda}. Xod would

answer tu'o, John would answer na'i. Same difference.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 13:59 GMT

pc:

> I apologize for getting sidetracked on xorxes' red herring here. The point

> is that {lo jvugi} (and {la jvugis} for that matter) is not transparent to

> conjunction.

My point was that {lo jvugi} behaves in the same way as {la jvugis}.

What I think is not very transparent to conjunction are hidden

assumptions. Saying:

la jvugis cu nenri [ca le cabdei] gi'enai nenri [ca le purlamdei]

"Jvugi is inside [today] and not inside [yesterday]."

without the bracketed terms is as perverse as saying:

lo jvugi cu nenri [bu'u le zunle] gi'enai nenri [bu'u le pritu]

"Jvugis are inside [on my left] and not inside [on my right]."

I don't know whether such hidden assumptions are strictly

permitted or not, but in any case they could only be used as

riddles and not for normal communication.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 15:40 GMT

But of course these "hidden assumptions" are just what a generic usage — which is what we purport yo be dealing with — drop out (or, better, explicitly refuse to raise). If we have to bring them back in to make some (unused) rule work, then we have failed to cover the case at hand. we don't need transparency and normal languages (including logic) don't have it, so why go through all the metaphysical mumbo-jumbo to get it in (especially since even with it the trick doesn't solve the "problems").

Jorge Llambías wrote:

pc:

> I apologize for getting sidetracked on xorxes' red herring here. The point

> is that {lo jvugi} (and {la jvugis} for that matter) is not transparent to

> conjunction.

My point was that {lo jvugi} behaves in the same way as {la jvugis}.

What I think is not very transparent to conjunction are hidden

assumptions. Saying:

la jvugis cu nenri [ca le cabdei] gi'enai nenri [ca le purlamdei]

"Jvugi is inside [today] and not inside [yesterday]."

without the bracketed terms is as perverse as saying:

lo jvugi cu nenri [bu'u le zunle] gi'enai nenri [bu'u le pritu]

"Jvugis are inside [on my left] and not inside [on my right]."

I don't know whether such hidden assumptions are strictly

permitted or not, but in any case they could only be used as

riddles and not for normal communication.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 15:57 GMT

John E Clifford wrote:

>I say leave {lo} as it is and learn to say what you mean.

>

Yet weeks ago you expressed the need for "somewhat longer cmavo list" to

deal with Existence.

>But of course these "hidden assumptions" are just what a generic usage — which is what we purport yo be dealing with — drop out (or, better, explicitly refuse to raise). If we have to bring them back in to make some (unused) rule work, then we have failed to cover the case at hand. we don't need transparency and normal languages (including logic) don't have it, so why go through all the metaphysical mumbo-jumbo to get it in (especially since even with it the trick doesn't solve the "problems").

>

>

I think that Jorge is saying that we already have temporal transparency,

so why not permit spatial? I see symmetry. And there is no prohibition

against using the detailed expressions in either case.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Mon 07 of June, 2004 17:36 GMT

A> And your point is? There are some things which Lojban does not say very precisely and it would be nice to have devices for dealing with them. A new {lo} is not on that list however.

B> But in the relevant sense we do not have even temporal transparency, so — if they are analogous — we should not expect spatial either. The point remains about generic claims, not about specifications of those generic claims, which are likely to be different, depending on how they are specified.

xod wrote:

John E Clifford wrote:

>I say leave {lo} as it is and learn to say what you mean.

>

A>Yet weeks ago you expressed the need for "somewhat longer cmavo list" to

deal with Existence.

>But of course these "hidden assumptions" are just what a generic usage — which is what we purport yo be dealing with — drop out (or, better, explicitly refuse to raise). If we have to bring them back in to make some (unused) rule work, then we have failed to cover the case at hand. we don't need transparency and normal languages (including logic) don't have it, so why go through all the metaphysical mumbo-jumbo to get it in (especially since even with it the trick doesn't solve the "problems").

>

>

B>I think that Jorge is saying that we already have temporal transparency,

so why not permit spatial? I see symmetry. And there is no prohibition

against using the detailed expressions in either case.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Mon 07 of June, 2004 17:45 GMT

Here's my longer response to Nora's No-vote post.

On Sun, Jun 06, 2004 at 05:27:36PM -0700, [email protected] wrote:

> I am voting "no" at this point. This does not mean I am unalterably

> opposed. The volume of traffic has been too much for me to keep up

> with to be able to make a real decision.

I don't consider a vote, pro or con, made without reading the proposal

to be made in good faith. In fact, I consider it obstructionist, as I

have tried to make clear.

I am hereby requesting, as jatna, that you withdraw your unfounded vote

(or, rather, request me to do so, as there is no way to remove votes

without directly altering the database).

You should know, by the way, that yours is the No vote from a BPFK

member (the other No vote is from PC, who is not a member), and as such

will not be enough to stop the passage of this checkpoint, even if your

vote *was* a legal vote under the BPFK procedures.

> I think that the original 2-week time-period was unrealistic,

> especially for this controversial a topic. In addition, after any

> changes or significant volume of discussion, there should be at least

> 2 weeks to re-review everthing; this will also give some leeway for

> people, like me, who don't spend all day every day reading e-mail or

> the tiki. There have been 28 messages since Robin posted the latest

> extension.

I resent the implication that I, or John Cowan, or Mark Shoulson, or any

of the other thirteen people that have voted on this are shiftless

layabouts that have nothing better to do with their time than read

e-mail.

Beyond that, however, two weeks after each change is absolutely out of

the question, and here's why:

There are sixty-six sections in the BPFK. I have been trying to not do

more then 3 or 4 per-checkpoint, but let's be generous. Let's say each

checkpoint averages out to 4 sections. I think in practice it's going

to be more like 2, but let's go with 4. That's approximately 16

checkpoints.

It seems to take about two weeks for a checkpoint to get into a form

that is worth voting on. At that point, we wrangle on it for another

two weeks and, when consensus among those active on the topic has been

reached, I close voting because hey, the goal is consensus.

At that rate, we're looking at *six*-*teen* *months* of BPFK work, of

which we've done *three*. That's another year and change of this.

Your suggestion would add another *eigth* *months* to the proccess, at

an absolutely bare minimum, and with very generous (to you) assumptions.

I will not support any such idea.

-Robin


Constants, quantifiers


Posted by rab.spir on Mon 07 of June, 2004 17:45 GMT posts: 152

I hadn't noticed that the thing about constants was part of the proposal.

Is it absolutely essential to the proposal that {lo broda} is a constant? The concept doesn't make very much sense to me, while the idea that {lo} is a generic article makes perfect sense. The constants thing seems like an unwanted rider on the proposal to me.

I would also prefer if the outer quantifier for {lo} had the same meaning as the one for {le}. But to help me understand it better, in the current proposal:

How do you say "the five groups of four students"?

How do you say "two of the five groups of four students"?

How do you say "the teacher will choose two out of five groups of four students"?

Basically, can you nest quantifiers arbitrarily, and how does it come out?

And when someone makes a post here without leaving a signature line, how do you tell who it is from the e-mail?

- Rob



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

Rob:

> Constants, quantifiers

> I hadn't noticed that the thing about constants was part of the proposal.

>

> Is it absolutely essential to the proposal that {lo broda} is a constant?

Yes. At least to the same extent that {la broda} and {le broda} are

constants.

> The

> concept doesn't make very much sense to me, while the idea that {lo} is a

> generic article makes perfect sense. The constants thing seems like an

> unwanted rider on the proposal to me.

All it means is that:

lo broda naku brode

is the same as

naku lo broda cu brode

just as with {la broda} and {le brode}.

Also, you can use a pronoun with {lo broda} as its antecedent, even

from outside of its prenex, just as with {la broda} and {le broda}.

> I would also prefer if the outer quantifier for {lo} had the same meaning as

> the one for {le}. But to help me understand it better, in the current

> proposal:

>

> How do you say "the five groups of four students"?

le mu lo vo tadni

> How do you say "two of the five groups of four students"?

re le mu lo vo tadni

> How do you say "the teacher will choose two out of five groups of four

> students"?

le ctuca ba cuxna re da lo mu lo vo tadni

> Basically, can you nest quantifiers arbitrarily, and how does it come out?

Yes, as you would expect.

> And when someone makes a post here without leaving a signature line, how do

> you tell who it is from the e-mail?

You can't, you have to go to the wiki page.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

On Mon, Jun 07, 2004 at 10:45:13AM -0700, Rob Speer? wrote:

> I hadn't noticed that the thing about constants was part of the

> proposal.

It is part of the Notes. As such, it is descriptive and annotative, but

largely subordinate to the definitions themselves.

> Is it absolutely essential to the proposal that {lo broda} is a

> constant?

Looking from a predicate logic perpective, I am told that it is. I care

very little about that aspect of it.

> The concept doesn't make very much sense to me, while the idea that

> {lo} is a generic article makes perfect sense. The constants thing

> seems like an unwanted rider on the proposal to me.

Ignore it, then. I certainly am for the most part.

> I would also prefer if the outer quantifier for {lo} had the same

> meaning as the one for {le}.

It is not very useful to talk about "two arbitrary, unspecified members

out of an arbitrary, unspecified three-some of broda". This is *not*

the same thing as "two out of every three broda", in the "two out of

three doctors prefer SnakeOil brand painkiller" sense.

The latter, IIRC, is "re fi'u ci lo broda".

> But to help me understand it better, in the current proposal:

>

> How do you say "the five groups of four students"?

Depends on what you mean by "the".

le mu lo vo selctu

is what I think you mean: The thing have in mind and am describing as

five groups arbitrary groups of four arbitrary students each.

> How do you say "two of the five groups of four students"?

Assuming I had you right the first time:

re le mu lo vo selctu

> How do you say "the teacher will choose two out of five groups of four

> students"?

Again, depending on context and "the":

lo ctuca .ei cu cuxna re le mu lo vo selctu

or, probably better:

lo ctuca .ei cu cuxna re girzu be fi mu lo'i vo selctu

> Basically, can you nest quantifiers arbitrarily, and how does it come

> out?

Does that help?

> And when someone makes a post here without leaving a signature line, how

> do you tell who it is from the e-mail?

>

> - Rob

You don't. That's why I keep badgering people to do it. I'll try to

fix the code at some point.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

On Mon, Jun 07, 2004 at 11:00:41AM -0700, Jorge Llamb?as wrote:

> > How do you say "the teacher will choose two out of five groups of four

> > students"?

>

> le ctuca ba cuxna re da lo mu lo vo tadni

Forgot about the place structure of cuxna. Ignore mine; this is better.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

IMNSHO, the last part of the definition for lo'e, starting at "It has been

proposed that ...", should be moved to the Notes section. The definitions

must be maximally prescriptive.

--

Arnt Richard Johansen http://arj.nvg.org/

The savvy DXer will usually bet a pair of nickels he can pinpoint the DXCC

country of the "chopity-chow-pit chow-chow-pi-chow" even before he hears the

call sign. -John F. Lindholm, QST vol. 66 no. 3 p. 83



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

On Mon, Jun 07, 2004 at 08:30:56PM +0200, Arnt Richard Johansen wrote:

> IMNSHO, the last part of the definition for lo'e, starting at "It has

> been proposed that ...", should be moved to the Notes section. The

> definitions must be maximally prescriptive.

I agree; this is pretty clearly a usage convention.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT


> On Mon, Jun 07, 2004 at 08:30:56PM +0200, Arnt Richard Johansen wrote:

> > IMNSHO, the last part of the definition for lo'e, starting at "It has

> > been proposed that ...", should be moved to the Notes section. The

> > definitions must be maximally prescriptive.

>

> I agree; this is pretty clearly a usage convention.

OK. I also removed the note that started "Personally, I think..."

because I don't think people should be voting on what I should

think.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 19:01 GMT

On Mon, Jun 07, 2004 at 11:40:19AM -0700, Jorge Llamb?as wrote:

>

> --- Robin Lee Powell wrote:

> > On Mon, Jun 07, 2004 at 08:30:56PM +0200, Arnt Richard Johansen

> > wrote:

> > > IMNSHO, the last part of the definition for lo'e, starting at "It

> > > has been proposed that ...", should be moved to the Notes section.

> > > The definitions must be maximally prescriptive.

> >

> > I agree; this is pretty clearly a usage convention.

>

> OK. I also removed the note that started "Personally, I think..."

> because I don't think people should be voting on what I should think.

Fair enough. Thanks.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT

On Sun, Jun 06, 2004 at 09:56:56PM -0400, John Cowan wrote:

> Jorge Llamb?as scripsit:

>

> > What is the status of the "Notes" section? I understand it as not

> > being a part of the definitions we're voting on, but just for

> > comments and clarifications. Robin?

>

> In that case, I'd like this to be added to the definitions of loi and

> lei:

>

> (A substance may be treated as a group made up of individuals

> without worrying about which individuals they are.)

Why not in the notes?

-Robin



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT


> On Sun, Jun 06, 2004 at 09:56:56PM -0400, John Cowan wrote:

> > In that case, I'd like this to be added to the definitions of loi and

> > lei:

> >

> > (A substance may be treated as a group made up of individuals

> > without worrying about which individuals they are.)

>

> Why not in the notes?

I have now added it as:

"A substance can also be seen as made up of component parts,

and this can even be true in a physical sense: the water in a

puddle of rain got there drop by drop. If you look from far

enough away, you can't tell whether the "sailor all over the

deck" is literally goo, or just an aggregation of sailors

standing still. So a substance may be treated as a group

made up of individuals without worrying about which individuals

they are."

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT

Jorge Llamb?as scripsit:

> OK. I also removed the note that started "Personally, I think..."

> because I don't think people should be voting on what I should

> think.

Good. I have three more changes to propose:

a) Alter the inner quantifiers of lai and la'i to agree with la, so lai ci cribe

is the Three Bears rather than the three Bears.

b) Alter the outer quantifier of loi to agree with lo and lo'i.

c) Clarify that an integral outer quantifier does not make sense with lei

unless an explicit inner quantifier is given. If change a is rejected,

this change should be extended to lai as well.

I have prepared a table at http://www.ccil.org/~cowan/gadri.html which

summarizes the proposal as it now stands in tabular form. Feel free to

incorporate this into your proposal; I will keep altering it to track

whatever changes you make.

--

John Cowan [email protected] www.ccil.org/~cowan www.reutershealth.com

"If I have seen farther than others, it is because I am surrounded by dwarves."

--Murray Gell-Mann



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT

Robin Lee Powell scripsit:

> > (A substance may be treated as a group made up of individuals

> > without worrying about which individuals they are.)

>

> Why not in the notes?

That was conditional on the notes not being part of the proposal

--

John Cowan [email protected] www.reutershealth.com www.ccil.org/~cowan

If a traveler were informed that such a man as Lord John Russell was

leader of the House of Commons, he may well begin to comprehend how the

Egyptians worshiped an insect. --Benjamin Disraeli



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT

A>But it isn't. The second denies {lo broda cu brode} so is true even if there are no broda, while {lo broda naku brode} is not.

B> But one must be very careful about what anaphoric pronouns we use. To be sure, Lojban — just like English — does not yet make the careful distinction needed. The difference generally does not make a difference for {la} and {le}, which are as close to constants as we are likely to find.

C>This is one possibility; but not the most likely in the context in which these examples arose.

Jorge Llambías wrote:

Rob:

> Constants, quantifiers

> I hadn't noticed that the thing about constants was part of the proposal.

>

> Is it absolutely essential to the proposal that {lo broda} is a constant?

Yes. At least to the same extent that {la broda} and {le broda} are

constants.

> The

> concept doesn't make very much sense to me, while the idea that {lo} is a

> generic article makes perfect sense. The constants thing seems like an

> unwanted rider on the proposal to me.

A>All it means is that:

lo broda naku brode

is the same as

naku lo broda cu brode

just as with {la broda} and {le brode}.

B>Also, you can use a pronoun with {lo broda} as its antecedent, even

from outside of its prenex, just as with {la broda} and {le broda}.

> I would also prefer if the outer quantifier for {lo} had the same meaning as

> the one for {le}. But to help me understand it better, in the current

> proposal:

>

> How do you say "the five groups of four students"?

le mu lo vo tadni

> How do you say "two of the five groups of four students"?

re le mu lo vo tadni

> How do you say "the teacher will choose two out of five groups of four

> students"?

C> le ctuca ba cuxna re da lo mu lo vo tadni

> Basically, can you nest quantifiers arbitrarily, and how does it come out?

Yes, as you would expect.

> And when someone makes a post here without leaving a signature line, how do

> you tell who it is from the e-mail?

You can't, you have to go to the wiki page.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 20:35 GMT

pc:

> But it isn't. The second denies {lo broda cu brode} so is true even if

> there are no broda, while {lo broda naku brode} is not.

As long as the same distinction applies to {la broda naku brode}

vs. {naku la broda cu brode}, then there's no problem.

But when there are broda, the two are the same. That does not

apply to {su'o broda}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT

A> But {lo broda naku brode} implies {su'o broda naku brode}, while {naku lo broda cu brode} does not.

Jorge Llambías wrote:

pc:

> But it isn't. The second denies {lo broda cu brode} so is true even if

> there are no broda, while {lo broda naku brode} is not.

As long as the same distinction applies to {la broda naku brode}

vs. {naku la broda cu brode}, then there's no problem.

A>But when there are broda, the two are the same. That does not

apply to {su'o broda}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT

On Mon, Jun 07, 2004 at 04:17:46PM -0400, [email protected]

wrote:

> Jorge Llamb?as scripsit:

>

> > OK. I also removed the note that started "Personally, I think..."

> > because I don't think people should be voting on what I should

> > think.

>

> Good. I have three more changes to propose:

>

> a) Alter the inner quantifiers of lai and la'i to agree with la, so

> lai ci cribe is the Three Bears rather than the three Bears.

>

> b) Alter the outer quantifier of loi to agree with lo and lo'i.

>

> c) Clarify that an integral outer quantifier does not make sense with

> lei unless an explicit inner quantifier is given. If change a is

> rejected, this change should be extended to lai as well.

For the record, I agree with all of these.

> I have prepared a table at http://www.ccil.org/~cowan/gadri.html which

> summarizes the proposal as it now stands in tabular form. Feel free

> to incorporate this into your proposal; I will keep altering it to

> track whatever changes you make.

He also has http://home.ccil.org/~cowan/gadri-cll.html, which is an

attempt at doing a similar table for what's currently in the CLL.

I'd like to see one or both of these in the proposal (presumably in the

Notes). I will do the conversion to Tiki tables if you like.

-Robin



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT

pc:

> But {lo broda naku brode} implies {su'o broda naku brode}, while {naku lo

> broda cu brode} does not.

When there are broda? What would be an example?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT


> On Mon, Jun 07, 2004 at 04:17:46PM -0400, [email protected]eutershealth.com

> wrote:

> > Good. I have three more changes to propose:

> >

> > a) Alter the inner quantifiers of lai and la'i to agree with la, so

> > lai ci cribe is the Three Bears rather than the three Bears.

> >

> > b) Alter the outer quantifier of loi to agree with lo and lo'i.

> >

> > c) Clarify that an integral outer quantifier does not make sense with

> > lei unless an explicit inner quantifier is given. If change a is

> > rejected, this change should be extended to lai as well.

>

> For the record, I agree with all of these.

I have incorporated a) and b).

About c): lei/le'i/lai/la'i would be the only remaining cases of

outer quantifiers not doing quantification. I'm willing to go

whichever way people like with these, but the most harmonious

with the system, I think, would be quantification over members

for lei/lai and quantification over subsets for le'i/la'i.

(Not that quantifying over subsets is especially useful, I can't

really think of interesting uses, but neither can I think of

interesting uses for the other interpretation.)

Comments?

> > I have prepared a table at http://www.ccil.org/~cowan/gadri.html which

> > summarizes the proposal as it now stands in tabular form. Feel free

> > to incorporate this into your proposal; I will keep altering it to

> > track whatever changes you make.

>

> He also has http://home.ccil.org/~cowan/gadri-cll.html, which is an

> attempt at doing a similar table for what's currently in the CLL.

I added links.

> I'd like to see one or both of these in the proposal (presumably in the

> Notes). I will do the conversion to Tiki tables if you like.

We could add it once we're done, if you want.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT

On Mon, Jun 07, 2004 at 02:22:07PM -0700, Jorge Llamb?as wrote:

> > John:

> Robin:

> > > c) Clarify that an integral outer quantifier does not make sense

> > > with lei unless an explicit inner quantifier is given. If change

> > > a is rejected, this change should be extended to lai as well.

> >

> > For the record, I agree with all of these.

>

> I have incorporated a) and b).

>

> About c): lei/le'i/lai/la'i would be the only remaining cases of outer

> quantifiers not doing quantification.

You mean besides le and la?

> I'm willing to go whichever way people like with these, but the most

> harmonious with the system, I think, would be quantification over

> members for lei/lai and quantification over subsets for le'i/la'i.

I know it would make xod very happy to have things be consistent one way

or the other.

I'm not inclined to comment as I still don't understand the difference

between "quantifying over members of the group" and "indicating a

subgroup".

I know this is a lot of work, but could you please translate the

following sentences using each version (i.e. first with what's currently

in the proposal, and secondly with what you're suggesting)?

re lei ci broda

re le'i ci broda

ci lei re broda

ci le'i re broda

re lai ci broda

re la'i ci broda

ci lai re broda

ci la'i re broda

-Robin



Posted by Anonymous on Mon 07 of June, 2004 22:42 GMT

Oops! Rather {su'o broda naku brode} implies a> {lo broda naku brode}, but not b> {naku lo broda cu brode}, since the a may be true together with {lo broda cu brode}.

Jorge Llambías wrote:

pc:

> But {lo broda naku brode} implies {su'o broda naku brode}, while {naku lo

> broda cu brode} does not.

When there are broda? What would be an example?

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:43 GMT

Jorge Llamb?as scripsit:

> About c): lei/le'i/lai/la'i would be the only remaining cases of

> outer quantifiers not doing quantification. I'm willing to go

> whichever way people like with these, but the most harmonious

> with the system, I think, would be quantification over members

> for lei/lai and quantification over subsets for le'i/la'i.

The difficulty there is that quantifying over members presumes that

there are members to quantify over, and sometimes we no longer remember

what they are.

> (Not that quantifying over subsets is especially useful, I can't

> really think of interesting uses, but neither can I think of

> interesting uses for the other interpretation.)

I think quantifying over subsets is bogus, because it makes all the

difference in the world which subset it is. "Two subsets of the set

of Beatles"? Bah.

--

John Cowan [email protected] www.reutershealth.com www.ccil.org/~cowan

The known is finite, the unknown infinite; intellectually we stand

on an islet in the midst of an illimitable ocean of inexplicability.

Our business in every generation is to reclaim a little more land,

to add something to the extent and the solidity of our possessions.

--Thomas Henry Huxley



Posted by Anonymous on Mon 07 of June, 2004 22:43 GMT


> On Mon, Jun 07, 2004 at 02:22:07PM -0700, Jorge Llamb?as wrote:

> > About c): lei/le'i/lai/la'i would be the only remaining cases of outer

> > quantifiers not doing quantification.

>

> You mean besides le and la?

Outer quantifiers on {le/la broda} quantify over members of the group

you have in mind and which you describe/name with broda. (That has

not changed from CLL.)

> I'm not inclined to comment as I still don't understand the difference

> between "quantifying over members of the group" and "indicating a

> subgroup".

"indicating a subgroup" is my way of saying "I'm not quite sure

what's going on, but I'm sure it is not quantification".

> I know this is a lot of work, but could you please translate the

> following sentences using each version (i.e. first with what's currently

> in the proposal, and secondly with what you're suggesting)?

>

> re lei ci broda

Now: Probably equivalent to {lo re le ci broda}

Suggestion: {re le ci broda}

> re le'i ci broda

Now: Probably {lo'i re le ci broda} i.e. a set with two

of the three brodas as members.

Suggestion: Two subsets of le'i broda.

> ci lei re broda

Now: meaningless

Suggestion: meaningless

> ci le'i re broda

Now: meaningless

Suggestion: Three subsets of le'i re broda.

(A set with two members happens to have four subsets.)

> re lai ci broda

> re la'i ci broda

> ci lai re broda

> ci la'i re broda

Analogously to lei/le'i.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:43 GMT


> Jorge Llamb?as scripsit:

>

> > About c): lei/le'i/lai/la'i would be the only remaining cases of

> > outer quantifiers not doing quantification. I'm willing to go

> > whichever way people like with these, but the most harmonious

> > with the system, I think, would be quantification over members

> > for lei/lai and quantification over subsets for le'i/la'i.

>

> The difficulty there is that quantifying over members presumes that

> there are members to quantify over, and sometimes we no longer remember

> what they are.

Right, but the other choice also requires members.

The choice seems to be between:

PA1 lei PA2 broda = lo PA1 le PA2 broda

PA1 lei PA2 broda = PA1 le PA2 broda

The second one just duplicates {le}. The first one turns the outer

quantifier into a sort of inner quantifier. I doubt this will

see much use either way.

> > (Not that quantifying over subsets is especially useful, I can't

> > really think of interesting uses, but neither can I think of

> > interesting uses for the other interpretation.)

>

> I think quantifying over subsets is bogus, because it makes all the

> difference in the world which subset it is. "Two subsets of the set

> of Beatles"? Bah.

It is not meaningless, but I definitely agree it has practically

no use. At least I don't see one. The other options not only have

no use, but also break the system.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 22:43 GMT

pc:

> Oops! Rather {su'o broda naku brode} implies a> {lo broda naku brode}, but

> not b> {naku lo broda cu brode}, since the a may be true together with {lo

> broda cu brode}.

{lo broda naku brode} can be true together with {lo broda cu brode} only

in the sense that {la broda naku brode} can be true together with

{la broda cu brode}, i.e. at different unmentioned times/places/

circumstances. But if everything else is kept constant, they can't

be true together.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 23:40 GMT

On Mon, Jun 07, 2004 at 03:02:04PM -0700, Jorge Llamb?as wrote:

> --- Robin Lee Powell wrote:

> > I'm not inclined to comment as I still don't understand the

> > difference between "quantifying over members of the group" and

> > "indicating a subgroup".

>

> "indicating a subgroup" is my way of saying "I'm not quite sure what's

> going on, but I'm sure it is not quantification".

Aaaah.

> > I know this is a lot of work, but could you please translate the

> > following sentences using each version (i.e. first with what's

> > currently in the proposal, and secondly with what you're

> > suggesting)?

> >

> > re lei ci broda

>

> Now: Probably equivalent to {lo re le ci broda}

>

> Suggestion: {re le ci broda}

That makes it identical to le except for emphasis, correct?

> > re le'i ci broda

>

> Now: Probably {lo'i re le ci broda} i.e. a set with two

> of the three brodas as members.

>

> Suggestion: Two subsets of le'i broda.

Which two?

> > ci le'i re broda

>

> Now: meaningless

>

> Suggestion: Three subsets of le'i re broda.

> (A set with two members happens to have four subsets.)

> > re lai ci broda

> > re la'i ci broda

> > ci lai re broda

> > ci la'i re broda

How is that possible if the number is part of the name?

-Robin



Posted by Anonymous on Mon 07 of June, 2004 23:40 GMT

Robin:

> > > re lei ci broda

> > Now: Probably equivalent to {lo re le ci broda}

> > Suggestion: {re le ci broda}

>

> That makes it identical to le except for emphasis, correct?

Yes.

> > > re le'i ci broda

> > Now: Probably {lo'i re le ci broda} i.e. a set with two

> > of the three brodas as members.

> > Suggestion: Two subsets of le'i broda.

>

> Which two?

Quantification never specifies which.

> > > re lai ci broda

> > > re la'i ci broda

> > > ci lai re broda

> > > ci la'i re broda

>

> How is that possible if the number is part of the name?

The thing can still have members, even though you can't be sure that

the number of members always agrees with the number in the name you

give it.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 23:58 GMT

Oh, give it a rest. The point is that the inference you claim holds does not in fact hold. The reason why it does not hold is that the particular quantifier — which regardless of how often you yack on the contrary is all that {lo} is or ever was — does not distribute the way you want it to. The whole point of generic claims is precisely that they do not involve particular times and places etc,.so, whatever may be true about claims that do involve particular times and places is not relevant. To be sure, finding out about the truth of these claims may involve looking at some particular times and places, but they are not part of the claim. The spurious analogy with individuals fails because of the larger disanalogy — there is the individual in the world and there is not Mr. Rabbit.

[email protected]> wrote:

pc:

> Oops! Rather {su'o broda naku brode} implies a> {lo broda naku brode}, but

> not b> {naku lo broda cu brode}, since the a may be true together with {lo

> broda cu brode}.

{lo broda naku brode} can be true together with {lo broda cu brode} only

in the sense that {la broda naku brode} can be true together with

{la broda cu brode}, i.e. at different unmentioned times/places/

circumstances. But if everything else is kept constant, they can't

be true together.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Mon 07 of June, 2004 23:58 GMT

On Mon, Jun 07, 2004 at 04:54:52PM -0700, John E Clifford wrote:

> Oh, give it a rest.

If that irony was hot, you could cook eggs on it.

-Robin



Posted by Anonymous on Tue 08 of June, 2004 00:34 GMT

On Mon, Jun 07, 2004 at 04:58:35PM -0700, [email protected] wrote:

> John:

> > I have prepared a table at http://www.ccil.org/~cowan/gadri.html

> > which summarizes the proposal as it now stands in tabular form.

>

> Maybe the key phrases for masses should be:

>

> fit and collectively satisfy

> described as and collectively satisfy

> named with and collectively satisfy

>

> and for sets:

>

> has members that fit

> has members described as

> has members named with

Those last seem to imply that the set could contain other things. That

would be, at best, counterintutive.

lo'i prenu cu vasru lo mlatu

-Robin



Posted by Anonymous on Tue 08 of June, 2004 00:34 GMT


> On Mon, Jun 07, 2004 at 04:58:35PM -0700, [email protected] wrote:

> > and for sets:

> >

> > has members that fit

> > has members described as

> > has members named with

>

> Those last seem to imply that the set could contain other things. That

> would be, at best, counterintutive.

Ok then:

has only members that fit

has only members described as

has only members named with

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 00:34 GMT

And the phrases for lo'e and le'e:

Those that fit typically satisfy

Those described as are said to typically satisfy

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 00:34 GMT

I personally made all the jvugi. There are exactly 12 of them, some of them are inside my box, some of them are outside. So, {lo jvugi cu nenri le tanxe} is true as is {lo jvugi naku nenri le tanxe}, at the same time and in the same place and so (as long as they are outside the box it doesn't matter, they are not at the only relevant place, inside the box). But I {lo jvugi cu jvugi} is true, so one of these jvugi must be both inside and outside the box, which it clearly is not. But these are all the jvugi. So, at least one of the lojban sentences above is false, if {lo jvugi} is a constant (as it is not as already noted). Which will you give up — the ones which are there in front of you or one of the ones that is supported only by a (to be generous) dubious metaphysics.

Jorge Llambías wrote:

pc:

> Oops! Rather {su'o broda naku brode} implies a> {lo broda naku brode}, but

> not b> {naku lo broda cu brode}, since the a may be true together with {lo

> broda cu brode}.

{lo broda naku brode} can be true together with {lo broda cu brode} only

in the sense that {la broda naku brode} can be true together with

{la broda cu brode}, i.e. at different unmentioned times/places/

circumstances. But if everything else is kept constant, they can't

be true together.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

pc:

> I personally made all the jvugi. There are exactly 12 of them, some of them

> are inside my box, some of them are outside. So, {lo jvugi cu nenri le

> tanxe} is true as is {lo jvugi naku nenri le tanxe}, at the same time and in

> the same place

But surely it can't be true at the exact same spot. Maybe in the same room,

on the same table, but not at the exact same spot. If all the jvugi are

superimposed one on top of another then they are all in or all out of

the box.

I personally made la jvugis, which scintillates so that any given second

it is emitting and not emitting light many times. However, if you look

at a very precise point in time, it is either emitting or not emitting,

but never both.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

This is getting really boring., I point out a simple logical truth to you about an argument and you tell me something about another argument. The you take a dubious claim about space and support it with a slightly less dubious claim about time, as though they were somehow the same. The only thing that differentiates the subjects in the various sentences is that, as different occurrences of the particular quantifier, they may — and in this case do — point to different jvugi. Since no place or time or anything else is mentioned here, only the separate quantifiers, what do place and time (just what "generic" is all about ignoring) have to do with the issue?

And we still have the question of which of these claims are you going to give up, because you can't have them all.

As for the analogy between the temporal slices of an individual and the spatially separated instnaces of a generality. it just doesn't work to prove that the instances are all phases of the same indivviual: individuals (if we must get metaphysical) are, if concrete, spatially and temporally continuous.

But, as I have said before, if you really think you want a constant here (but I have yet to see a real problem that such a thing would solve) there is at least species (and probably a few others of that ilk) that you can use. The thing talk about species is always idiomatic, that is, it is never literally true but is a shorthand for species talk. But it will — under that condition — do whatever you want a constant to do, even converge over a conjunction of contradictory properties (the species overlaps both P and not-P, that is, is subsunmed by neither), There are some problems with empirical truth and existence here, since — as it goes now — Lojban doesn't distinguish between what is true conceptually and what empirically. But pretty much the same cmavo that Lojban needs to do a better job with the standard {lo} could be used to solve that instead. Unfortunetely, even species don't pass through negation unharmed (and there is no time/place/instance/whatever else you might

think would help to fall back on, species being the sort of things they are, nodes on the semantic net).

But again, why bother with all of this (though I think species may be something we want to deal with eventually)? What can one of these things do (even assuming it could do all the things you claim for it) that needs to be done (speaker sloth — and malgliconess — aside) that old {lo} can't?

And, of course, by using these odd things you manage to disguise again at least one of the more interest features which old {lo} turned up, the ambiguity of the notion of anaphora and the corresponding ambiguity of (English) pronouns (now projected onto Lojban and so in need of correcting there).

Jorge Llambías wrote:

pc:

> I personally made all the jvugi. There are exactly 12 of them, some of them

> are inside my box, some of them are outside. So, {lo jvugi cu nenri le

> tanxe} is true as is {lo jvugi naku nenri le tanxe}, at the same time and in

> the same place

But surely it can't be true at the exact same spot. Maybe in the same room,

on the same table, but not at the exact same spot. If all the jvugi are

superimposed one on top of another then they are all in or all out of

the box.

I personally made la jvugis, which scintillates so that any given second

it is emitting and not emitting light many times. However, if you look

at a very precise point in time, it is either emitting or not emitting,

but never both.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

Taking your purported argument somewhat more seriously than it deserves, let us get away from generics and consider the situation here and now (in however narrow or broad a sense you will. With some jvugi in the box (here) and others not, we have as true {lo jvugi ca vi nenri le tanxe} and {lo jvugi naku ca vi nenri le tanxe}, but not (since the first is true) {naku lo jvugi ca vi nenri le tanxe}.

Jorge Llambías wrote:

pc:

> I personally made all the jvugi. There are exactly 12 of them, some of them

> are inside my box, some of them are outside. So, {lo jvugi cu nenri le

> tanxe} is true as is {lo jvugi naku nenri le tanxe}, at the same time and in

> the same place

But surely it can't be true at the exact same spot. Maybe in the same room,

on the same table, but not at the exact same spot. If all the jvugi are

superimposed one on top of another then they are all in or all out of

the box.

I personally made la jvugis, which scintillates so that any given second

it is emitting and not emitting light many times. However, if you look

at a very precise point in time, it is either emitting or not emitting,

but never both.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

pc:

> Taking your purported argument somewhat more seriously than it deserves, let

> us get away from generics and consider the situation here and now (in however

> narrow or broad a sense you will. With some jvugi in the box (here) and

> others not, we have as true {lo jvugi ca vi nenri le tanxe} and {lo jvugi

> naku ca vi nenri le tanxe}, but not (since the first is true) {naku lo jvugi

> ca vi nenri le tanxe}.

What happens with la jvugis in the analogous set up? If la jvugis

is here and now emitting and not emitting light, then

{la jvugis ca zi te gusni} and {la jvugis ca zi naku te gusni}, but

not (since the first is true) {naku la jvugis ca zi te gusni}.

But it is not {la jvugis} or {lo jvugi} that are interfering with

{naku}, it is {vi} and {zi}, both of which contain points where ja'aku

and points where naku.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

The point is, of course, that, ex hypothesi, la jvugis is an individual and so cannot be both emitting and not emitting light here and now (which may — to make you happy — require defining here and now rather narrowly, to be sure). As for the latter point, note where the negation is in my example — already outside the tense place. Again I have made a point about one argument and you have countered with a point about another — irrelevant — argument. Trying to defend an obvious lunacy does that to one eventually.

Jorge Llambías wrote:

pc:

> Taking your purported argument somewhat more seriously than it deserves, let

> us get away from generics and consider the situation here and now (in however

> narrow or broad a sense you will. With some jvugi in the box (here) and

> others not, we have as true {lo jvugi ca vi nenri le tanxe} and {lo jvugi

> naku ca vi nenri le tanxe}, but not (since the first is true) {naku lo jvugi

> ca vi nenri le tanxe}.

What happens with la jvugis in the analogous set up? If la jvugis

is here and now emitting and not emitting light, then

{la jvugis ca zi te gusni} and {la jvugis ca zi naku te gusni}, but

not (since the first is true) {naku la jvugis ca zi te gusni}.

But it is not {la jvugis} or {lo jvugi} that are interfering with

{naku}, it is {vi} and {zi}, both of which contain points where ja'aku

and points where naku.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

Jorge Llamb?as scripsit:

> It is not meaningless, but I definitely agree it has practically

> no use. At least I don't see one. The other options not only have

> no use, but also break the system.

Fair enough. In that case, it might be better just to say that non-fractional

quantifiers are simply undefined.

--

John Cowan www.ccil.org/~cowan www.reutershealth.com

Micropayment advocates mistakenly believe that efficient allocation of

resources is the purpose of markets. Efficiency is a byproduct of market

systems, not their goal. The reasons markets work are not because users

have embraced efficiency but because markets are the best place to allow

users to maximize their preferences, and very often their preferences are

not for conservation of cheap resources. --Clay Shirkey



Posted by Anonymous on Tue 08 of June, 2004 13:34 GMT

pc:

> The point is, of course, that, ex hypothesi, la jvugis is an individual and

> so cannot be both emitting and not emitting light here and now (which may --

> to make you happy — require defining here and now rather narrowly, to be

> sure).

lo jvugi cannot be both inside and not inside the box here and now

for a narrow "here" either.

> As for the latter point, note where the negation is in my example --

> already outside the tense place.

Right, I missed that. In that case, I disagree that {lo jvugi naku

ca vi nenri} is true with a vi that includes the jvugis inside the box.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 14:24 GMT

John Cowan:

> Fair enough. In that case, it might be better just to say that

> non-fractional

> quantifiers are simply undefined.

The way I think outer fractional quantifiers should work,

non-fractionals are a special case of fractionals (with

{ro} as denominator) so defining how fractionals work

automatically defines non-fractionals.

So I gather your preference for lei and le'i is for the

status quo, which I interpret as:

PA lei broda = lo PA le broda

PA le'i broda = lo'i PA le broda

where PA can be a fractional.

So for example:

pa fi'u re lei broda

= lo pa fi'u re le broda

= A group consisting of one out of every two

of the group of broda I have in mind.

= A group of half of the broda I have in mind.

pa fi'u re le'i broda

= lo'i pa fi'u re le broda

= A set of one out of every two of the broda I have in mind.

= A set of half of the broda I have in mind

= A set with half the members of the set of broda I have in mind.

I understand that's what the current definitions say.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 14:25 GMT

A> How do you do that? There are clearly jvugi that are not here and now inside the box: thye are not here and they are not inside the box, although they are now, but 2 out of 3 (indeed 1 out of 3) is enough to decide the negation. Do you understand your own idea of what {lo jvigu} means? Or {naku}?

In any case, what will you end up proving. So far we have established that linguistic proper names are not logical constants, which is hardly a surprise since 20th century British analytic philosophy started with Russell making that point (rathe more tidily than we have but still involving negation transparency essentially). The counterinstances are cases that look paradoxical: Jvugi is glowing and Jvugi is not glowing. So, to resolve the paradox, we artificially divide the unit subject, Jvugi, into parts, so that we can say "part of Jvugi is glowing and part of him is not." The parts involved have no independent reality beyond the analytic needs of the moment. So supposed analogy to this is {lo broda cu brede ije lo broda naku brode}. But — in official Lojban — this is not even paradoxical, let alone needing some cure. But, to make it paradoxical (are there other reasons?), we artifically create Mr. Broda, referred to by all {lo broda}s. Like the parts of the person created

by analysis, this creature of synthesis has no independent reality. Since all it has managed to do so far is create an apparent paradox out of a perfectly ordinary claim (and it should be noted that the claim involving {lo} can be perfectly ordinary, while the problem with names involve a lot of stage setting before we can get the right kind of case).

wrote:

pc:

> The point is, of course, that, ex hypothesi, la jvugis is an individual and

> so cannot be both emitting and not emitting light here and now (which may --

> to make you happy — require defining here and now rather narrowly, to be

> sure).

lo jvugi cannot be both inside and not inside the box here and now

for a narrow "here" either.

> As for the latter point, note where the negation is in my example --

> already outside the tense place.

A>Right, I missed that. In that case, I disagree that {lo jvugi naku

ca vi nenri} is true with a vi that includes the jvugis inside the box.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 15:54 GMT

pc:

> So far we have established that

> linguistic proper names are not logical constants, which is hardly a surprise

> since 20th century British analytic philosophy started with Russell making

> that point (rathe more tidily than we have but still involving negation

> transparency essentially).

That's why I say that {lo jvugi} is a constant to the same extent

that {la jvugis} is.

> Since

> all it has managed to do so far is create an apparent paradox out of a

> perfectly ordinary claim (and it should be noted that the claim involving

> {lo} can be perfectly ordinary, while the problem with names involve a lot of

> stage setting before we can get the right kind of case).

{su'o broda cu brode ije su'o broda naku brode} is as clear and

ordinary as "Some brodas brode and some brodas don't brode".

{lo broda cu brode ije lo broda naku brode} is as paradoxical (or not)

as "Brodas brode and brodas don't brode".

There is no need to identify the two.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT

Or{la jvugis} fails to be a constant to the extent that {lo jvugi} does. But you seem to want to say that {lo jvugi} is more of a constant that {la jvugis}, since you seem to want to say tha there is some sense in which {lo jvugi} collects over conjunction and passes unscathed through negation in a way that {la jvugis} does not. The most that can be said accurately about la jvugis in a collective way is that some part of it shines and some (other, of course) part of it does not. {la jvugis zu'o ge pisu'o jy te gusni gi pisu'o jy naku te gusni}, but you deny the corresponding form for {lo jvugi}. And well you should, because the it has to admit quantifers over {lo jvugi} which gets back the old meaning of {lo} and explains at least this case in terms of them ({lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni} — unlless instances are parts too, in which case I suppose the exact parallel would work). To which the immediate rhetorical reply would be, if you need them

already, why bother with the added bit that doesn't help (and indeed created the paradox in the first place).

B> From my point of view not paradoxical at all. But that is because {lo broda} simply refers to different broda in the two places. If they refer to one and the same broda, then there is an apparent paradox and we solve it by saying — in your terms — some instances of lo broda do and some don't.

C> Whjich two? The sentences? There may be no need to identify them but they mutually entail one another and so aare materially equivalent. That is, there is no reason to separate them.

Jorge Llambías wrote:

pc:

> So far we have established that

> linguistic proper names are not logical constants, which is hardly a surprise

> since 20th century British analytic philosophy started with Russell making

> that point (rathe more tidily than we have but still involving negation

> transparency essentially).

A>That's why I say that {lo jvugi} is a constant to the same extent

that {la jvugis} is.

> Since

> all it has managed to do so far is create an apparent paradox out of a

> perfectly ordinary claim (and it should be noted that the claim involving

> {lo} can be perfectly ordinary, while the problem with names involve a lot of

> stage setting before we can get the right kind of case).

{su'o broda cu brode ije su'o broda naku brode} is as clear and

ordinary as "Some brodas brode and some brodas don't brode".

B>{lo broda cu brode ije lo broda naku brode} is as paradoxical (or not)

as "Brodas brode and brodas don't brode".

C>There is no need to identify the two.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT

pc:

> lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni

That's perfectly fine. It makes no sense to replace {lo jvugi}

with {su'o jvugi} there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT


>

> pc:

> > lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni

>

> That's perfectly fine. It makes no sense to replace {lo jvugi}

> with {su'o jvugi} there.

In fact, you need {boi}s between su'o and jy, but other than

that, it's perfectly fine.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT

Well, why not? To be sure, just what {su'o su'o jvugi} means is probably unclear and (alternately) what requantifying variables means is controversial. Or is your point just that it is stylistically inelegant? Perhaps; I never said there were no differences between {su'o} and {lo} ({lo PA brode} for one simple case). But to say that it makes *no* sense is just false.

Jorge Llambías wrote:

pc:

> lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni

That's perfectly fine. It makes no sense to replace {lo jvugi}

with {su'o jvugi} there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT

Presumably in the {la jvugis} case too. {boi} is one of those patches that I have never quite figued out how it works nor why it is there in the first place, but I'll take your word that it is needed here.

Jorge Llambías wrote:


>

> pc:

> > lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni

>

> That's perfectly fine. It makes no sense to replace {lo jvugi}

> with {su'o jvugi} there.

In fact, you need {boi}s between su'o and jy, but other than

that, it's perfectly fine.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:27 GMT

It had better mean something (and something like {su'o jvigu cu nenri le tanxe ije su'o broda naku nenri le tanxe} because it follows from the sentence given, which is then senseless if the {suo} for {lo} replacement is. The alternative is that your {lo jvigu} can't do the job it is designed to do, say what jvigu do.

pc:

> lo jvugi zo'u ge su'o jy te gusni gi su'o jy naku te gusni

That's perfectly fine. It makes no sense to replace {lo jvugi}

with {su'o jvugi} there.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:28 GMT

pc:

> {boi} is one of those patches that I

> have never quite figued out how it works nor why it is there in the first

> place, but I'll take your word that it is needed here.

The problem is that digits and lerfu always merge, so {su'o by}

is a number. When the time to discuss grammar changes comes, I

am proposing that digits and lerfu be disentangled, so that

{su'o by} works as one would expect.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:28 GMT

Another example:

lo jvugi zo'u roboi jy broda ije su'oboi jy brode

ije noboi jy brodi ije me'iboi jy brodo

"As for jvugis, all of them broda, some of them brode,

none of them brodi and not all of them brodo."

I wouldn't use {su'o jvugi} in the prenex instead of {lo jvugi}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:28 GMT

Good!

Jorge Llambías wrote:

pc:

> {boi} is one of those patches that I

> have never quite figued out how it works nor why it is there in the first

> place, but I'll take your word that it is needed here.

The problem is that digits and lerfu always merge, so {su'o by}

is a number. When the time to discuss grammar changes comes, I

am proposing that digits and lerfu be disentangled, so that

{su'o by} works as one would expect.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:28 GMT

I am sure you wouldn't and I wouldn't either, but for stylistic-- maybe even grammatical — reasons. As I said before, requantifying is a controversy, but suely {ro lo jvigu} or {ro jvigu} (whichever {jy} stands for here). The initial {lo jvigu} is vacuous, of course, or downright misleading, but not wrong.

Jorge Llambías wrote:

Another example:

lo jvugi zo'u roboi jy broda ije su'oboi jy brode

ije noboi jy brodi ije me'iboi jy brodo

"As for jvugis, all of them broda, some of them brode,

none of them brodi and not all of them brodo."

I wouldn't use {su'o jvugi} in the prenex instead of {lo jvugi}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 21:28 GMT

pc:

> I am sure you wouldn't and I wouldn't either, but for stylistic-- maybe even

> grammatical — reasons. As I said before, requantifying is a controversy,

> but suely {ro lo jvigu} or {ro jvigu} (whichever {jy} stands for here).

{jy} stands for {lo jvigu} here, which is unquantified.

> The

> initial {lo jvigu} is vacuous, of course, or downright misleading, but not

> wrong.

Let's change to a non-vacuous one then:

la pycyn cu nelci lo nu zbasu lo jvugi i roboi jy broda

ije su'oboi jy brode ije noboi jy brodi ije me'iboi jy brodo

"pc likes making jvugis. All of them broda, some of them brode,

none of them brodi and not all of them brodo."

I wouldn't use {su'o jvugi} there instead of {lo jvugi}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 22:22 GMT

Yes, this use is non-vacuous and the quantifiers in the subsequent sentences only quantify over the one I make (which is, in fact, all of them of course, so the distinction isn't worth much — but that is fact, not logic). As far as I can tell, {nelci} does not create opaque places, so the pronouns work just fine. The questions about {jy} and requantification do keep coming back, but — without deciding those issues completely — this seems to be right. I am not sure just what this has to do with {lo}, but I can't complain about the sentences.

Jorge Llambías wrote:

pc:

> I am sure you wouldn't and I wouldn't either, but for stylistic-- maybe even

> grammatical — reasons. As I said before, requantifying is a controversy,

> but suely {ro lo jvigu} or {ro jvigu} (whichever {jy} stands for here).

{jy} stands for {lo jvigu} here, which is unquantified.

> The

> initial {lo jvigu} is vacuous, of course, or downright misleading, but not

> wrong.

Let's change to a non-vacuous one then:

la pycyn cu nelci lo nu zbasu lo jvugi i roboi jy broda

ije su'oboi jy brode ije noboi jy brodi ije me'iboi jy brodo

"pc likes making jvugis. All of them broda, some of them brode,

none of them brodi and not all of them brodo."

I wouldn't use {su'o jvugi} there instead of {lo jvugi}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Tue 08 of June, 2004 22:22 GMT

But on second thought — this one devoted to anaphora, but not requantification (for which I am at least temporarily assuming that the each subsequent quantifier subdivides the original one — not the latest one and not back to scratch either) — I want the anaphora after the first sentence to be referential, not literal. But I have (not quite arbitrarily) said that the letterals are literal, so I would use other pronouns here. Unfortunately, once we get rid of the letterals, Lojban — for all it has reams of sheets of pronouns — does not have any that relaibly pick out something already mentioned or pick it out again after one shot. so, for the nonce, I'll let {jy} be as ambiguous as the English cases.

John E Clifford wrote:Yes, this use is non-vacuous and the quantifiers in the subsequent sentences only quantify over the one I make (which is, in fact, all of them of course, so the distinction isn't worth much — but that is fact, not logic). As far as I can tell, {nelci} does not create opaque places, so the pronouns work just fine. The questions about {jy} and requantification do keep coming back, but — without deciding those issues completely — this seems to be right. I am not sure just what this has to do with {lo}, but I can't complain about the sentences.

Jorge Llambías wrote:

pc:

> I am sure you wouldn't and I wouldn't either, but for stylistic-- maybe even

> grammatical — reasons. As I said before, requantifying is a controversy,

> but suely {ro lo jvigu} or {ro jvigu} (whichever {jy} stands for here).

{jy} stands for {lo jvigu} here, which is unquantified.

> The

> initial {lo jvigu} is vacuous, of course, or downright misleading, but not

> wrong.

Let's change to a non-vacuous one then:

la pycyn cu nelci lo nu zbasu lo jvugi i roboi jy broda

ije su'oboi jy brode ije noboi jy brodi ije me'iboi jy brodo

"pc likes making jvugis. All of them broda, some of them brode,

none of them brodi and not all of them brodo."

I wouldn't use {su'o jvugi} there instead of {lo jvugi}.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 00:35 GMT

pc:

> Yes, this use is non-vacuous and the quantifiers in the subsequent sentences

> only quantify over the one I make (which is, in fact, all of them of course,

> so the distinction isn't worth much — but that is fact, not logic).

Ok, so here is a difference in our understanding. Consider this variation:

lo jvugi le tanxe cu nenri gi'e bartu

i la pycyn cu nelci lo nu zbasu jy i roboi jy broda

ije su'oboi jy brode ije noboi jy brodi ije me'iboi jy brodo

"Jvugis are inside the box and out. pc likes making them.

All of them broda, some of them brode, none of them brodi

and not all of them brodo."

jy is always lo jvugi, whatever you have said about them or

how you have quantified over the instances.

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 00:35 GMT

Jorge "Llamb����������������������������������" wrote:

Mozilla 1.6 mailer. Go figure.

>pc:

>

>

>>Taking your purported argument somewhat more seriously than it deserves, let

>>us get away from generics and consider the situation here and now (in however

>>narrow or broad a sense you will. With some jvugi in the box (here) and

>>others not, we have as true {lo jvugi ca vi nenri le tanxe} and {lo jvugi

>>naku ca vi nenri le tanxe}, but not (since the first is true) {naku lo jvugi

>>ca vi nenri le tanxe}.

>>

>>

The reason why all this works (or doesn't) is that {lo jvugi} can refer

to more than one jvugi, and that's the problem for pc, since as a

"constant" it shouldn't be allowed to switch around like that. I

think. This discussion is well past me and I probably shouldn't be

talking on it.

>What happens with la jvugis in the analogous set up? If la jvugis

>is here and now emitting and not emitting light, then

>{la jvugis ca zi te gusni} and {la jvugis ca zi naku te gusni}, but

>not (since the first is true) {naku la jvugis ca zi te gusni}.

>

>

It's not as obvious, since {la jvugis} is *usually* taken to mean just

one thing, but it *can* mean more than one. Does that

help/hinder/otherwise affect thinking on the problem?

~mark



Posted by Anonymous on Wed 09 of June, 2004 00:35 GMT

Mark E. Shoulson:

> The reason why all this works (or doesn't) is that {lo jvugi} can refer

> to more than one jvugi, and that's the problem for pc, since as a

> "constant" it shouldn't be allowed to switch around like that. I

> think. This discussion is well past me and I probably shouldn't be

> talking on it.

I think pc would not like it even if we limit ourselves to

{lo pa jvugi}, i.e. singleton jvugies.

For example:

pamai lo jvugi cu xanri la pycyn i remai pycy skicu jy

i romai pycy zbasu jy

"First jvugis were imagined by pc, then he described them,

and fianlly he made them."

How many were imagined? How many were described? na'i!

(We know he made 12 though.)

> It's not as obvious, since {la jvugis} is *usually* taken to mean just

> one thing, but it *can* mean more than one. Does that

> help/hinder/otherwise affect thinking on the problem?

la jvugis can be a group, yes, but it is still a single referent.

In a given use it has a fixed referent, unlike variables.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 01:28 GMT

On Tuesday 08 June 2004 18:59, Jorge "Llamb=EDas" wrote:

> pc:

> > Yes, this use is non-vacuous and the quantifiers in the subsequent

> > sentences only quantify over the one I make (which is, in fact, all of

> > them of course, so the distinction isn't worth much — but that is fact,

> > not logic).

>

> Ok, so here is a difference in our understanding. Consider this variation:

>

> lo jvugi le tanxe cu nenri gi'e bartu

{lo jvugi le tanxe cu nenri gi'e bartu} is false.

{lo jvugi le tanxe cu nenri gi'a bartu} is true.

{lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true (some =

are=20

in, some are out, but they're not the same ones).

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 01:59 GMT

Pierre Abbat:

> {lo jvugi le tanxe cu nenri gi'e bartu} is false.

All else being equal, yes, but:

lo jvugi le tanxe cu pu nenri gi'e ca bartu

or:

lo jvugi le tanxe cu va nenri gi'e vi bartu

can be true. Jvugis are inside the box over there and outside over here.

> {lo jvugi le tanxe cu nenri gi'a bartu} is true.

Almost tautologically true :-)

> {lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true

> (some are in, some are out, but they're not the same ones).

This is the expansion of the first one.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Minor re-wording.


rlpowell Posted by rlpowell on Wed 09 of June, 2004 02:00 GMT posts: 14214

In loi and lei:

and which satisfy the predicate for which the sumti is an argument collectively

should be

and which collectively satisfy the predicate for which the sumti is an argument

-Robin



Posted by Anonymous on Wed 09 of June, 2004 02:09 GMT

I haven't been following this (I stopped getting mail about bpfk

stuff a long time ago---I think it was moved away from the TWiki

or something like that), and because of that I'm not entirely sure

whether I'm going to vote.

First thing, so that the rest of this isn't misunderstood: the

lojban gadri system is definitely not complete currently---this is

probably one of the main things which has made me lose a lot of

interest in the language. I agree that changes are needed.

However....

I think the proposed change to {lo} invalidates substatial portions

of CLL and *lots* of past usage (although correct usage of {lo} was

admittedly somewhat rare). For some reason it isn't being advertised

as such on its page. Moreover it's a hacky way to add generics to

the language. In many cases the current {loi} works for sentences

which you want a generic article for (check out the example sentence

for {loi} on the wiki page). And if a new function is actually

necessary, it would be much cleaner to use {lau} for it than to do

something like this, which complicates the meaning of {lo} based

on whether its quantifiers are explicit.

How it changes the meaning of things:

lojbo: .ei lo verba cu mutce fraxu lo makcu prenu

xorlo: Children should show great forbearance toward grown-up people.

old-lo: There is a child for whom there is an adult, where the former

should show great forbearance toward the latter.

lojbo: ca lo nicte lo cinfo cu kalte lo cidja

xorlo: At night lions hunt for food.

old-lo: There was a night when there was a lion and some food;

the former was hunting for the latter.

lojbo: lo pavyseljirna cu ranmi danlu gi'e simlu

lo ka ge ce'u xirma gi lo pa jirna cu cpana lo mebri be ce'u

xorlo: Unicorns are mythical creatures that look like a horse with a

horn coming out of their foreheads.

old-lo: There is a unicorn who is a mythical animal, and who looks like

a horse with a/the horn which is the only horn in existence

upon one of its heads.

Note also that under the old lo, this uses the nonsensical

"lo ka". All {ka} or {du'u} (which are the same thing) are

"le", by virtue of the fact that the speaker is about to

say it.

There are also examples of real usage that are invalidated, aside

from the above example sentences. At this point I'm less concerned

about the effects things have on existing usage, since most existing

usage is wrong anyway. But on IRC Robin seemed to be of the opinion

that this change to {lo} did not invalidate existing usage, which

does not seem to me to be the case:

My translation of Carver's "Little Things":

lojbo: do bersa lo malpre

xorlo: You are a son of bitches (in general).

Or more englishy: Bitches are your mother.

old-lo: There is a bitch who is your mother.

Your mother's a bitch.

Nick's aesop stuff:

lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

xorlo: Foxes are hungry and when they see some grapes

dangling from trees want to take them but can't.

old-lo: A fox, who was hungry, upon seeing some grapes in a tree

wanted to take them, but couldn't.

>From my translation of Ambrose Bierce's "The Man and the Snake":

lojbo: lo since cu nenri lo sipna kumfa

xorlo: Snakes are inside of bedrooms.

old-lo: There is a snake for whom there is a bedroom containing him.

In the context of: {le'enu lo since cu nenri lo sipna kumfa

be lo cabna ke tcadu zdani noi zmadu cu to xamgu toi na

dukse leka fadni kei lenu na jai se sarcu tu'a lo ve ciksi}

a

One other thing I'd like to point out is that currently {lo} is the

only article that is defined in terms of the lower-level predicate

logic features of the language (lo broda == da poi broda). First

of all this really sucks---more of the language should be that

explicitly defined if lojban really is supposed to be a "logical

language".

But regarding this---why try to kill the only article which *is*

well defined, instead of fixing the various ill-defined articles

(i.e. there is no definite consensus on which properties of individual

members a mass gets and which properties emerge, and we need an

article for the old problem of accidently declaring the existance

of imaginary things (a la {mi kalte LE pavyseljirna}), so that we

don't need to keep working around it with tanru).

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxnOBDrrilS51AZ8RAoTRAKCTAG3GzxkxxEIomaAqQrmtoNx9wwCfbTA7

yf1rO2K7VXJwgz7XmPORrUM=

=z/JH

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 02:35 GMT

On Tuesday 08 June 2004 21:51, Jorge "Llamb=EDas" wrote:

> Pierre Abbat:

> > {lo jvugi le tanxe cu nenri gi'e bartu} is false.

>

> All else being equal, yes, but:

>

> lo jvugi le tanxe cu pu nenri gi'e ca bartu

That can be true, if a jvugi was taken out of the box, ...

> or:

>

> lo jvugi le tanxe cu va nenri gi'e vi bartu

>

> can be true. Jvugis are inside the box over there and outside over here.

but I don't see how this can be true, unless {nenri} and {bartu} are true a=

t=20

different times.

> > {lo jvugi le tanxe cu nenri gi'a bartu} is true.

>

> Almost tautologically true :-)

>

> > {lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true

> > (some are in, some are out, but they're not the same ones).

>

> This is the expansion of the first one.

It looks like it, but {lo jvugi} can have a different referent each time it=

=20

occurs, whereas in the first one {lo jvugi} occurs once, so the sentence is=

=20

claiming that a jvugi is both inside and outside the box.

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 02:35 GMT

> should be

> and which collectively satisfy the predicate for which the sumti is an

> argument

> -Robin

Done. I have also added a summary table based on John's to the page.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 02:43 GMT


> On Tuesday 08 June 2004 21:51, Jorge "Llamb=EDas" wrote:

> >

> > lo jvugi le tanxe cu va nenri gi'e vi bartu

>

> but I don't see how this can be true, unless {nenri} and {bartu} are

> true at different times.

It can be true at the same time because {lo jvugi} is not "some jvugi"

but jvugis in general.

> > > {lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true

> > > (some are in, some are out, but they're not the same ones).

> >

> > This is the expansion of the first one.

>

> It looks like it, but {lo jvugi} can have a different referent each

> time it occurs,

It always has the same referent: jvugis.

> whereas in the first one {lo jvugi} occurs once, so the sentence

> claiming that a jvugi is both inside and outside the box.

It just claims that jvugis are both inside and outside. It doesn't

talk about instances.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 03:02 GMT

Looks fine to me; all the {jy} are functioning as literal anaphora, though I suppose we disagree on that. We agree both on what the grammar is and what it means but not how it means that. And, of course, both of our understandings are in their own ways non-standard.

Jorge Llambías wrote:

pc:

> Yes, this use is non-vacuous and the quantifiers in the subsequent sentences

> only quantify over the one I make (which is, in fact, all of them of course,

> so the distinction isn't worth much — but that is fact, not logic).

Ok, so here is a difference in our understanding. Consider this variation:

lo jvugi le tanxe cu nenri gi'e bartu

i la pycyn cu nelci lo nu zbasu jy i roboi jy broda

ije su'oboi jy brode ije noboi jy brodi ije me'iboi jy brodo

"Jvugis are inside the box and out. pc likes making them.

All of them broda, some of them brode, none of them brodi

and not all of them brodo."

jy is always lo jvugi, whatever you have said about them or

how you have quantified over the instances.

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Nitpicky errors


Posted by noras on Wed 09 of June, 2004 03:02 GMT posts: 23

A couple of nit-picky errors (corrections welcome if I am mistaken):

Under "le" examples

"ko punji le sicni lo porsi be lo vamrai bi'o lo vamtolrai"

"Put the coins in order from greatest to least value."

The x3 of punji is the place where the x2 is put

Put the coins at place: sequence(s) from ... I think perhaps "ko zukte fi le nu le'i sicni cu te porsi lo vamrai bi'o lo vamtolrai", but this would then not be an example of "le". Or maybe "ko punji le sicni sepo'i lo vamrai bi'o lo vamtolrai".

Under "la" examples

"ma'i la midju terdi la sadam na sai me la sauron"

"In Middle Earth-terms, Saddam is by no means a 'Sauron'."

You really have 2 reference-frames wanted here: the real world and Middle Earth. Saddam is in one, and the Sauron is in the other. I read the lojban as setting the "world" to Middle Earth for the whole sentence, in which case "Saddam" is meaningless. The Middle Earth reference-frame applies only to 'Sauron'. So I suggest "la sadam na sai me la sauron pe ma'i la midju terdi" = "Saddam is by no means a 'Sauron'-of-middle-earth"

These are not all my comments. I am working on the rest.



Posted by Anonymous on Wed 09 of June, 2004 03:14 GMT

It is not the plurality of {lo jvugi} but its variability that is the problem. Or more exactly that it is a number things being treated simultaneously as though they were just one. The {la jvugis} case is actually made easier by plurality, but then it is even more obviously not a constant.

"Mark E. Shoulson" wrote:Jorge "Llamb����������������������������������" wrote:

Mozilla 1.6 mailer. Go figure.

>pc:

>

>

>>Taking your purported argument somewhat more seriously than it deserves, let

>>us get away from generics and consider the situation here and now (in however

>>narrow or broad a sense you will. With some jvugi in the box (here) and

>>others not, we have as true {lo jvugi ca vi nenri le tanxe} and {lo jvugi

>>naku ca vi nenri le tanxe}, but not (since the first is true) {naku lo jvugi

>>ca vi nenri le tanxe}.

>>

>>

The reason why all this works (or doesn't) is that {lo jvugi} can refer

to more than one jvugi, and that's the problem for pc, since as a

"constant" it shouldn't be allowed to switch around like that. I

think. This discussion is well past me and I probably shouldn't be

talking on it.

>What happens with la jvugis in the analogous set up? If la jvugis

>is here and now emitting and not emitting light, then

>{la jvugis ca zi te gusni} and {la jvugis ca zi naku te gusni}, but

>not (since the first is true) {naku la jvugis ca zi te gusni}.

>

>

It's not as obvious, since {la jvugis} is *usually* taken to mean just

one thing, but it *can* mean more than one. Does that

help/hinder/otherwise affect thinking on the problem?

~mark



Posted by Anonymous on Wed 09 of June, 2004 03:14 GMT

A> Well, I do think it is easier to stick with jvugis rather than singleytons of jvugi, but the problem turns up either way. If there were-- overtly — only one jvugi, of course, the situation would be somewhat simpler

B> We don't say of course; that is a large part of what "generic" means.

Jorge Llambías wrote:

Mark E. Shoulson:

> The reason why all this works (or doesn't) is that {lo jvugi} can refer

> to more than one jvugi, and that's the problem for pc, since as a

> "constant" it shouldn't be allowed to switch around like that. I

> think. This discussion is well past me and I probably shouldn't be

> talking on it.

A>I think pc would not like it even if we limit ourselves to

{lo pa jvugi}, i.e. singleton jvugies.

For example:

pamai lo jvugi cu xanri la pycyn i remai pycy skicu jy

i romai pycy zbasu jy

"First jvugis were imagined by pc, then he described them,

and fianlly he made them."

B>How many were imagined? How many were described? na'i!

(We know he made 12 though.)

> It's not as obvious, since {la jvugis} is *usually* taken to mean just

> one thing, but it *can* mean more than one. Does that

> help/hinder/otherwise affect thinking on the problem?

la jvugis can be a group, yes, but it is still a single referent.

In a given use it has a fixed referent, unlike variables.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 03:45 GMT

I haven't gotten this one from xorxes yet; but I haven't anything to add to our comments.

Pierre Abbat wrote:On Tuesday 08 June 2004 18:59, Jorge "Llamb=EDas" wrote: > pc:

> > Yes, this use is non-vacuous and the quantifiers in the subsequent

> > sentences only quantify over the one I make (which is, in fact, all of

> > them of course, so the distinction isn't worth much — but that is fact,

> > not logic).

>

> Ok, so here is a difference in our understanding. Consider this variation:

>

> lo jvugi le tanxe cu nenri gi'e bartu

{lo jvugi le tanxe cu nenri gi'e bartu} is false.

{lo jvugi le tanxe cu nenri gi'a bartu} is true.

{lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true (some =

are=20

in, some are out, but they're not the same ones).

phma

=2D-=20

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 03:45 GMT

coi fi'i ui doi jordan

> lojbo: .ei lo verba cu mutce fraxu lo makcu prenu

> xorlo: Children should show great forbearance toward grown-up people.

> old-lo: There is a child for whom there is an adult, where the former

> should show great forbearance toward the latter.

I think rather

oldlo: It ought to be that there is some child that shows great

forbearance toward some adult.

i.e. {ei} has scope over quantifiers, not the other way around.

> My translation of Carver's "Little Things":

> lojbo: do bersa lo malpre

> xorlo: You are a son of bitches (in general).

> Or more englishy: Bitches are your mother.

> old-lo: There is a bitch who is your mother.

> Your mother's a bitch.

Why not "you are the son of at least one bitch"? And why

"mother" rather than "parent"?

I don't see any problem with {do bersa lo malpre} with

the proposed lo.

> Nick's aesop stuff:

> lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

> xorlo: Foxes are hungry and when they see some grapes

> dangling from trees want to take them but can't.

> old-lo: A fox, who was hungry, upon seeing some grapes in a tree

> wanted to take them, but couldn't.

Why do you translate the second as a particular event and the first one

as a general event, given that there is no indication either way in

the Lojban? old-lo could just as well be: "some foxes are hungry

and when they see some grapes dangling from some trees want to

take them but can't." The proposed {lo} does not impose a generic

reading to the whole sentence, it is just not explicit as to instances.

All it says is: "Fox, hungry, when sees that grape hangs from tree, wants

get it but can't get". How you choose to fill the gaps must come from

context.

> From my translation of Ambrose Bierce's "The Man and the Snake":

>

> lojbo: lo since cu nenri lo sipna kumfa

> xorlo: Snakes are inside of bedrooms.

> old-lo: There is a snake for whom there is a bedroom containing him.

or: At least one snake is inside of at least one room.

With the proposed lo, I would simply say "snake inside room". How you

turn that into more fluent English will depend on the context. {lo}

adds nothing.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 03:46 GMT

A> True but not relevant to the question at issue, namely whether we could get Pierre's false claim from the true {lo jvugi le tanxe cu nenri ije lo jvugi le tanxe naku nenri}

B> Not really, since the first is not the collapse of it. The first is contradictory, the last is not.

Jorge Llambías wrote:

Pierre Abbat:

> {lo jvugi le tanxe cu nenri gi'e bartu} is false.

A>All else being equal, yes, but:

lo jvugi le tanxe cu pu nenri gi'e ca bartu

or:

lo jvugi le tanxe cu va nenri gi'e vi bartu

can be true. Jvugis are inside the box over there and outside over here.

> {lo jvugi le tanxe cu nenri gi'a bartu} is true.

Almost tautologically true :-)

> {lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true

> (some are in, some are out, but they're not the same ones).

B>This is the expansion of the first one.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 10:48 GMT

noras:

> "ko punji le sicni lo porsi be lo vamrai bi'o lo vamtolrai"

> "Put the coins in order from greatest to least value."

>

> The x3 of punji is the place where the x2 is put

> Put the coins at place: sequence(s) from ... I think perhaps "ko zukte fi

> le nu le'i sicni cu te porsi lo vamrai bi'o lo vamtolrai", but this would

> then not be an example of "le". Or maybe "ko punji le sicni sepo'i lo vamrai

> bi'o lo vamtolrai".

Right, how about:

ko punji le sicni ja'e lo porsi be lo vamrai bi'o lo vamtolrai

> "ma'i la midju terdi la sadam na sai me la sauron"

> "In Middle Earth-terms, Saddam is by no means a 'Sauron'."

>

> You really have 2 reference-frames wanted here: the real world and Middle

> Earth. Saddam is in one, and the Sauron is in the other. I read the lojban

> as setting the "world" to Middle Earth for the whole sentence, in which case

> "Saddam" is meaningless. The Middle Earth reference-frame applies only to

> 'Sauron'. So I suggest "la sadam na sai me la sauron pe ma'i la midju terdi"

> = "Saddam is by no means a 'Sauron'-of-middle-earth"

The predicate here is {me la sauron}. Don't BAI tags add a place to

the predicate, so that we have "x1 is a 'Sauron' by standard x"?

> These are not all my comments. I am working on the rest.

ki'e mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 10:49 GMT

On Tue, Jun 08, 2004 at 10:11:46PM -0400, Pierre Abbat wrote:

> On Tuesday 08 June 2004 21:51, Jorge "Llamb=EDas" wrote:

> > Pierre Abbat:

> > or:

> >

> > lo jvugi le tanxe cu va nenri gi'e vi bartu

> >

> > can be true. Jvugis are inside the box over there and outside over

> > here.

>

> but I don't see how this can be true, unless {nenri} and {bartu} are

> true at different times.

Just for the record, vi and va are place markers, not time markers.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 10:49 GMT

On Wednesday 09 June 2004 01:15, Robin Lee Powell wrote:

> On Tue, Jun 08, 2004 at 10:11:46PM -0400, Pierre Abbat wrote:

> > but I don't see how this can be true, unless {nenri} and {bartu} are

> > true at different times.

>

> Just for the record, vi and va are place markers, not time markers.

Yes. I am assuming that a jvugi can't be in two places at the same time.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 12:49 GMT

Pierre Abbat:

> I am assuming that a jvugi can't be in two places at the same time.

No jvugi can be in two places at once, but jvugis in general can

be in more than one place at the same time. Remember that {lo jvugi}

says nothing about the instances. All you say is "jvugi inside of

the box and outside of the box". How that relates to instances is

left unspecified. If you now say "at least one jvugi is both inside

and outside of the box at the same time" then I agree that's not

true. But that's not what {lo jvugi} says.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


Constants


Posted by xorxes on Wed 09 of June, 2004 14:10 GMT posts: 1912

Some comments on constant terms.

IF X and Z are constant terms, then the following equivalences

always hold:

X na broda == X naku broda == naku X broda == X broda naku

X broda Z == Z se broda X

X broda Z ije X brode Z == X boi Z broda gi'e brode

X broda lo nu Z brode ije ri brodi == X broda lo nu Z brode ije Z brodi

None of the above work in general with terms that carry an

outer quantifier.

With the proposal, every Lojban sumti with no explicit

outer quantifier is a constant (i.e. it carries no implicit quantifier)

and always permits the above transformations. (Even terms

like da behave that way when they are within the scope of

their quantifier and thus carry no outer quantifier. They

can't be manipulated like that when they do carry an outer

quantifier, you must first send the quantifier to a prenex.)

With CLL-defaults, {le broda} and {la broda} are pseudo-constants,

they carry a quantifier but they usually have a single referent, so

the transformations "usually" work. The proposal, by requiring

explicit quantifiers when quantification is intended, makes the

logic of terms much cleaner.

mu'o mi'e xorxes



Posted by Anonymous on Wed 09 of June, 2004 15:12 GMT

[email protected] scripsit:

> With the proposal, every Lojban sumti with no explicit

> outer quantifier is a constant (i.e. it carries no implicit quantifier)

> and always permits the above transformations. (Even terms

> like da behave that way when they are within the scope of

> their quantifier and thus carry no outer quantifier. They

> can't be manipulated like that when they do carry an outer

> quantifier, you must first send the quantifier to a prenex.)

Good.

Question about the semantics of "lo ci gerku cu batci le pa bongu":

does this entail that all of the dogs bit the bone, or is it left vague?

I'd prefer to require it to be all of them (as opposed to loi, where

they only have to bite it "collectively").

--

Do what you will, John Cowan

this Life's a Fiction [email protected]

And is made up of http://www.reutershealth.com

Contradiction. --William Blake http://www.ccil.org/~cowan



Posted by Anonymous on Wed 09 of June, 2004 15:12 GMT

On Tue, Jun 08, 2004 at 07:42:21PM -0700, Jorge Llamb?as wrote:

> It can be true at the same time because {lo jvugi} is not "some jvugi"

> but jvugis in general.

I don't care that this is discussion and not in the proposal, but if people

keep saying this, I will have to change my vote to NO until it's cleared up.

Why must {lo jvugi} refer to jvugis in general every time it's used? That

breaks past usage, and also is opposed to saying that the meaning of {lo} is

unspecified.

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

The official usage has its problems, too. {lo} is used too often on the assumption that just one thing is intended, where as it says that any number from one on is involved. So, where singularity is an issue, it is misleading. On the other hand, the new {lo} goes to the other extreme, getting away from single cases and single situations, while still (unremarked — even denied — by xorxes) staying within the scope of the particular quantifier. It is unclear to me whether this is a semantic issue or merely a pragmatic one: where do we place the (vague) limits on what counts as validating a claim? At the moment, I think it is pragmatic: the range is the same but we take it in one sense or the other: general or singular. Along that line, I have suggested (or was about to; I can't remember whether I have already said this in print) that we reserve {lo} henceforth for one of these assumptions and use {su'o} or a more exact number for the others. But I think that either of these will

run into a lot of conflicts with usage: xorxes' v. everybody else's.

While I object to xorxes reading of {lo} in some cases, in others it is clearly more nearly correct than the alternative suggested here.

<xorlo: Children should show great forbearance toward grown-up people.

old-lo: There is a child for whom there is an adult, where the former

should show great forbearance toward the latter>>

Clearly, first of all, the quantifiers are inside the opaque context of {ei}. Secondly, this is obviously not about any single child and single adult; it is meant generally (though not quite "all"). The second point applies (though less obviously) to the lion case, though there the single-interpretation is only improbable.

<lo ka ge ce'u xirma gi lo pa jirna cu cpana lo mebri be ce'u

xorlo: Unicorns are mythical creatures that look like a horse with a

horn coming out of their foreheads.

old-lo: There is a unicorn who is a mythical animal, and who looks like

a horse with a/the horn which is the only horn in existence

upon one of its heads.>>

This is a classic problem, since whatever we do (barring making {xanri1} opaque) we end up saying that there is something which is imaginary (a problem if we are quantifying only one this-world things). The "which is the only horn in existence" is not a problem about {lo} but a separate issue about what internal quantifiers mean (and, for now, I think xorxes has the right of it). I would have used {le mebri}, but "one of its heads" is pretty clearly not right as a translation, much less "at least one of its heads," which is correct but wrong. the comment about {le ka} seems right (even if there are several such properties, it is the one I am about to give that I mean). But equating {ka} and {du'u} will create some other problems down the line and fairly shortly (we will get back to this when the time comes).

The remaining (i.e., real) cases all seem to favor the "one" reading, but are, to that extent, as wrongheaded as the other reading. This comes out precisely because {lo} was well-defined: as "at least one...," not as "one" (and certainly not as "in general" though this is included). I agree that every gadri (and other cmavo eventually) needs a clear logical definition and I take it that ultimately that is what this BPFK exercise is about. Given that, the exercise has barely begun, let alone being over in the next week or two.

I see {kalte} is back as a problem: does something have to exist to be hunted in the {kalte} sense. Lifew is easier for logicians if we say no and make {kalte2} opaque. But it is harder for Lojbanists, since {kalte2} — like several other cases that have turned up lately — does not take an abstract argument and thus does not overtly provide a frame for other-world reference. Lojbanists have regularly rejectted the notion that some places are by themselves opaque. So, I guess that, whatever I may think, {mi kalte lo pavyseljirna} is false; I just think that I am doing that or what is really true is {mi pavyseljirna kalte} (the two stock philosphers' tricks here).

Jordan DeLong wrote:

I haven't been following this (I stopped getting mail about bpfk

stuff a long time ago---I think it was moved away from the TWiki

or something like that), and because of that I'm not entirely sure

whether I'm going to vote.

First thing, so that the rest of this isn't misunderstood: the

lojban gadri system is definitely not complete currently---this is

probably one of the main things which has made me lose a lot of

interest in the language. I agree that changes are needed.

However....

I think the proposed change to {lo} invalidates substatial portions

of CLL and *lots* of past usage (although correct usage of {lo} was

admittedly somewhat rare). For some reason it isn't being advertised

as such on its page. Moreover it's a hacky way to add generics to

the language. In many cases the current {loi} works for sentences

which you want a generic article for (check out the example sentence

for {loi} on the wiki page). And if a new function is actually

necessary, it would be much cleaner to use {lau} for it than to do

something like this, which complicates the meaning of {lo} based

on whether its quantifiers are explicit.

How it changes the meaning of things:

lojbo: .ei lo verba cu mutce fraxu lo makcu prenu

xorlo: Children should show great forbearance toward grown-up people.

old-lo: There is a child for whom there is an adult, where the former

should show great forbearance toward the latter.

lojbo: ca lo nicte lo cinfo cu kalte lo cidja

xorlo: At night lions hunt for food.

old-lo: There was a night when there was a lion and some food;

the former was hunting for the latter.

lojbo: lo pavyseljirna cu ranmi danlu gi'e simlu

lo ka ge ce'u xirma gi lo pa jirna cu cpana lo mebri be ce'u

xorlo: Unicorns are mythical creatures that look like a horse with a

horn coming out of their foreheads.

old-lo: There is a unicorn who is a mythical animal, and who looks like

a horse with a/the horn which is the only horn in existence

upon one of its heads.

Note also that under the old lo, this uses the nonsensical

"lo ka". All {ka} or {du'u} (which are the same thing) are

"le", by virtue of the fact that the speaker is about to

say it.

There are also examples of real usage that are invalidated, aside

from the above example sentences. At this point I'm less concerned

about the effects things have on existing usage, since most existing

usage is wrong anyway. But on IRC Robin seemed to be of the opinion

that this change to {lo} did not invalidate existing usage, which

does not seem to me to be the case:

My translation of Carver's "Little Things":

lojbo: do bersa lo malpre

xorlo: You are a son of bitches (in general).

Or more englishy: Bitches are your mother.

old-lo: There is a bitch who is your mother.

Your mother's a bitch.

Nick's aesop stuff:

lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

xorlo: Foxes are hungry and when they see some grapes

dangling from trees want to take them but can't.

old-lo: A fox, who was hungry, upon seeing some grapes in a tree

wanted to take them, but couldn't.

>From my translation of Ambrose Bierce's "The Man and the Snake":

lojbo: lo since cu nenri lo sipna kumfa

xorlo: Snakes are inside of bedrooms.

old-lo: There is a snake for whom there is a bedroom containing him.

In the context of: {le'enu lo since cu nenri lo sipna kumfa

be lo cabna ke tcadu zdani noi zmadu cu to xamgu toi na

dukse leka fadni kei lenu na jai se sarcu tu'a lo ve ciksi}

a

One other thing I'd like to point out is that currently {lo} is the

only article that is defined in terms of the lower-level predicate

logic features of the language (lo broda == da poi broda). First

of all this really sucks---more of the language should be that

explicitly defined if lojban really is supposed to be a "logical

language".

But regarding this---why try to kill the only article which *is*

well defined, instead of fixing the various ill-defined articles

(i.e. there is no definite consensus on which properties of individual

members a mass gets and which properties emerge, and we need an

article for the old problem of accidently declaring the existance

of imaginary things (a la {mi kalte LE pavyseljirna}), so that we

don't need to keep working around it with tanru).

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxnOBDrrilS51AZ8RAoTRAKCTAG3GzxkxxEIomaAqQrmtoNx9wwCfbTA7

yf1rO2K7VXJwgz7XmPORrUM=

=z/JH

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

On Wed, Jun 09, 2004 at 07:10:19AM -0700, [email protected] wrote:

> With the proposal, every Lojban sumti with no explicit

> outer quantifier is a constant (i.e. it carries no implicit quantifier)

> and always permits the above transformations. (Even terms

> like da behave that way when they are within the scope of

> their quantifier and thus carry no outer quantifier. They

> can't be manipulated like that when they do carry an outer

> quantifier, you must first send the quantifier to a prenex.)

>

> With CLL-defaults, {le broda} and {la broda} are pseudo-constants,

> they carry a quantifier but they usually have a single referent, so

> the transformations "usually" work. The proposal, by requiring

> explicit quantifiers when quantification is intended, makes the

> logic of terms much cleaner.

Yes, but apparently it does so by breaking past usage of {lo}.

Sorry, but this strikes me as one of the areas where {lo} is _supposed_ to be

different from {le} and {la} - it can refer to different things at different

times. Making {lo} generic is great. But if it's generic, it has to be able to

mean {su'o} in some contexts, and {su'o} is not a constant.

Sorry if I'm getting pissy. You were _so_ close to an entirely agreeable

proposal about {lo} that handled all past usage, but now constants are smeared

all over it.

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

A>But it implies (and is implied by {su'o jvugi} and so is false if the corresponding {su'o jvugi} is false.

B> See above.

C> See above.

Jorge Llambías wrote:


wrote:

> On Tuesday 08 June 2004 21:51, Jorge "Llamb=EDas" wrote:

> >

> > lo jvugi le tanxe cu va nenri gi'e vi bartu

>

> but I don't see how this can be true, unless {nenri} and {bartu} are

> true at different times.

A>It can be true at the same time because {lo jvugi} is not "some jvugi"

but jvugis in general.

> > > {lo jvugi le tanxe cu nenri .ije lo jvugi le tanxe cu bartu} is true

> > > (some are in, some are out, but they're not the same ones).

> >

> > This is the expansion of the first one.

>

> It looks like it, but {lo jvugi} can have a different referent each

> time it occurs,

B>It always has the same referent: jvugis.

> whereas in the first one {lo jvugi} occurs once, so the sentence

> claiming that a jvugi is both inside and outside the box.

C>It just claims that jvugis are both inside and outside. It doesn't

talk about instances.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

John Cowan:

> Question about the semantics of "lo ci gerku cu batci le pa bongu":

> does this entail that all of the dogs bit the bone, or is it left vague?

> I'd prefer to require it to be all of them (as opposed to loi, where

> they only have to bite it "collectively").

I don't think {lo ci gerku cu batci le pa bongu} says anything about

individual dogs. It says something about the one bone you have in mind

and dog trios. The bone gets bitten by dog trio. If you want to say

that three dogs bit the bone and each of them sank its teeth in, say

it: {ci gerku cu batci le pa bongu}. Your claim is then explicitly

about several instances of biting, and thus requires quantification.

{lo} can never fail if {loi} holds, because {loi} adds content. At

worst {lo} is more vague.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

D> See earlier.

Jorge Llambías wrote:

Pierre Abbat:

> I am assuming that a jvugi can't be in two places at the same time.

No jvugi can be in two places at once, but jvugis in general can

be in more than one place at the same time. Remember that {lo jvugi}

says nothing about the instances. All you say is "jvugi inside of

the box and outside of the box". How that relates to instances is

left unspecified. If you now say "at least one jvugi is both inside

and outside of the box at the same time" then I agree that's not

true. But that's not what {lo jvugi} says.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

D> See earlier.

Jorge Llambías wrote:

Pierre Abbat:

> I am assuming that a jvugi can't be in two places at the same time.

No jvugi can be in two places at once, but jvugis in general can

be in more than one place at the same time. Remember that {lo jvugi}

says nothing about the instances. All you say is "jvugi inside of

the box and outside of the box". How that relates to instances is

left unspecified. If you now say "at least one jvugi is both inside

and outside of the box at the same time" then I agree that's not

true. But that's not what {lo jvugi} says.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

On Tue, Jun 08, 2004 at 08:17:45PM -0700, Jorge Llambías wrote:

> coi fi'i ui doi jordan

> > lojbo: .ei lo verba cu mutce fraxu lo makcu prenu

> > xorlo: Children should show great forbearance toward grown-up people.

> > old-lo: There is a child for whom there is an adult, where the former

> > should show great forbearance toward the latter.

>

> I think rather

> oldlo: It ought to be that there is some child that shows great

> forbearance toward some adult.

> i.e. {ei} has scope over quantifiers, not the other way around.

Agreed.

> > My translation of Carver's "Little Things":

> > lojbo: do bersa lo malpre

> > xorlo: You are a son of bitches (in general).

> > Or more englishy: Bitches are your mother.

> > old-lo: There is a bitch who is your mother.

> > Your mother's a bitch.

>

> Why not "you are the son of at least one bitch"? And why

> "mother" rather than "parent"?

What I was translating from english. The lojban doesn't say anything

about gender deliberately, so it shouldn't be "mother" or "bitch"

in the back-translation to English, sorry.

...

> > Nick's aesop stuff:

> > lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> > dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

> > xorlo: Foxes are hungry and when they see some grapes

> > dangling from trees want to take them but can't.

> > old-lo: A fox, who was hungry, upon seeing some grapes in a tree

> > wanted to take them, but couldn't.

>

> Why do you translate the second as a particular event and the first one

> as a general event, given that there is no indication either way in

> the Lojban? old-lo could just as well be: "some foxes are hungry

> and when they see some grapes dangling from some trees want to

> take them but can't." The proposed {lo} does not impose a generic

> reading to the whole sentence, it is just not explicit as to instances.

> All it says is: "Fox, hungry, when sees that grape hangs from tree, wants

> get it but can't get". How you choose to fill the gaps must come from

> context.

The old {lo} does impose that it is a real, existing event, which

occurs for at least one fox, who saw at least one tree, etc.

Correct me if I'm wrong, but xorlo doesn't require that the fox

actually exists. It could work just as well for me to say:

lo pavyseljirna noi xagji ku'o ca lenu viska loi srasu noi

cpana lo rajma'a cu djica lenu cpacu ra gi'e naka'e cpacu

(Of course the above could work in old-lo, but only in a work of

fiction---no one will believe the speaker is talking about reality)

> > From my translation of Ambrose Bierce's "The Man and the Snake":

> >

> > lojbo: lo since cu nenri lo sipna kumfa

> > xorlo: Snakes are inside of bedrooms.

> > old-lo: There is a snake for whom there is a bedroom containing him.

>

> or: At least one snake is inside of at least one room.

Same thing. My gloss was intended to emphasize the logical structure

of it.

What it means is

Exx) . Ey(sipna_kumfa(y) . nenri(x, y?)

> With the proposed lo, I would simply say "snake inside room". How you

> turn that into more fluent English will depend on the context. {lo}

> adds nothing.

If you want that, why not say {since nenri kumfa}?

You didn't adress my point about this killing the only well-defined

article in the language. I think BPFK article changes should really

go towards further formalization, rather than away from the small

amount we currently have.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxyowDrrilS51AZ8RAvWoAJwL5J69W9b901Z/EW/i8fMatVnhrwCgoTex

TN0GuY5Ysb0w/jkF66c03r4=

=ffpV

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

A>As xorxes as demonstrated (though I don't think he thinks this is what he has demonstrated) none of these equations work for any sumti expression in Lojban. That is, lojban sumti of whatever sort are not logical constant. What we can do in some case — maybe in all — is rewrite some parts of these expressions to make the pattern go through — either divide the subject or precide the the predicate to avoid problems. The negation problems seem the hardest to deal with, but even there we can usually manage to rewrite what we say to achieve the shift for what we mean.

B> Is the proposal really this broad? If so, it is bound to fail in most cases, even with the "abstract" gadri like {lo'e}. Oddly enough, the case of {da} within the scope of its quantifier is the one case that pretty clearly is not problematic, since it is *guaranteed* a unique reference.

[email protected] wrote:

A>Constants

Some comments on constant terms.

IF X and Z are constant terms, then the following equivalences

always hold:

X na broda == X naku broda == naku X broda == X broda naku

X broda Z == Z se broda X

X broda Z ije X brode Z == X boi Z broda gi'e brode

X broda lo nu Z brode ije ri brodi == X broda lo nu Z brode ije Z brodi

None of the above work in general with terms that carry an

outer quantifier.

B>With the proposal, every Lojban sumti with no explicit

outer quantifier is a constant (i.e. it carries no implicit quantifier)

and always permits the above transformations. (Even terms

like da behave that way when they are within the scope of

their quantifier and thus carry no outer quantifier. They

can't be manipulated like that when they do carry an outer

quantifier, you must first send the quantifier to a prenex.)

With CLL-defaults, {le broda} and {la broda} are pseudo-constants,

they carry a quantifier but they usually have a single referent, so

the transformations "usually" work. The proposal, by requiring

explicit quantifiers when quantification is intended, makes the

logic of terms much cleaner.

mu'o mi'e xorxes



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

Rob Speer:

> > The proposal, by requiring

> > explicit quantifiers when quantification is intended, makes the

> > logic of terms much cleaner.

>

> Yes, but apparently it does so by breaking past usage of {lo}.

For example? Only if you were relying on the implicit quantifier

to transmit something significant is the usage broken. I don't

see many examples of that.

> Sorry, but this strikes me as one of the areas where {lo} is _supposed_ to be

> different from {le} and {la} - it can refer to different things at different

> times. Making {lo} generic is great. But if it's generic, it has to be able

> to

> mean {su'o} in some contexts, and {su'o} is not a constant.

{lo} allows that su'o is true, but it does not ever impose it.

Context may strongly suggest {su'o}, and the semantics of the

predicate may strongly require it (how can you see something

if you don't see at least one instance?) but the quantifier

is not part of the bare {lo} claim.

> Sorry if I'm getting pissy. You were _so_ close to an entirely agreeable

> proposal about {lo} that handled all past usage, but now constants are

> smeared

> all over it.

If you are not much concerned with the logic part of a

Lojban sentence you can disregard all talk about constants.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

A> Hell, under the present proposal, we just have that dog threesomes (in general) bite the one bone (which apparently moves around a lot to be available to dog threesomes wherever they are). The question of whether each member of every such threesome has to get its teeth around the bone seems relatively insignificant at this point.

John Cowan wrote:[email protected] scripsit:

> With the proposal, every Lojban sumti with no explicit

> outer quantifier is a constant (i.e. it carries no implicit quantifier)

> and always permits the above transformations. (Even terms

> like da behave that way when they are within the scope of

> their quantifier and thus carry no outer quantifier. They

> can't be manipulated like that when they do carry an outer

> quantifier, you must first send the quantifier to a prenex.)

Good.

A>Question about the semantics of "lo ci gerku cu batci le pa bongu":

does this entail that all of the dogs bit the bone, or is it left vague?

I'd prefer to require it to be all of them (as opposed to loi, where

they only have to bite it "collectively").

--

Do what you will, John Cowan

this Life's a Fiction [email protected]

And is made up of http://www.reutershealth.com

Contradiction. --William Blake http://www.ccil.org/~cowan



Posted by Anonymous on Wed 09 of June, 2004 15:13 GMT

Amen (except to note that this proposal has been around for several years in nearly this form, so it is a long time since it was close to being right).

Rob Speer wrote:On Wed, Jun 09, 2004 at 07:10:19AM -0700, [email protected] wrote:

> With the proposal, every Lojban sumti with no explicit

> outer quantifier is a constant (i.e. it carries no implicit quantifier)

> and always permits the above transformations. (Even terms

> like da behave that way when they are within the scope of

> their quantifier and thus carry no outer quantifier. They

> can't be manipulated like that when they do carry an outer

> quantifier, you must first send the quantifier to a prenex.)

>

> With CLL-defaults, {le broda} and {la broda} are pseudo-constants,

> they carry a quantifier but they usually have a single referent, so

> the transformations "usually" work. The proposal, by requiring

> explicit quantifiers when quantification is intended, makes the

> logic of terms much cleaner.

Yes, but apparently it does so by breaking past usage of {lo}.

Sorry, but this strikes me as one of the areas where {lo} is _supposed_ to be

different from {le} and {la} - it can refer to different things at different

times. Making {lo} generic is great. But if it's generic, it has to be able to

mean {su'o} in some contexts, and {su'o} is not a constant.

Sorry if I'm getting pissy. You were _so_ close to an entirely agreeable

proposal about {lo} that handled all past usage, but now constants are smeared

all over it.

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 16:50 GMT

E> See earlier

F> rabspir's comments suggest that he is deeply concerned about the logic of Lojban and that is why he finds your proposal so objectionable: it replaces reasonably clear logic with a total muddle (and to no purpose to boot).

Jorge Llambías wrote:

Rob Speer:

> > The proposal, by requiring

> > explicit quantifiers when quantification is intended, makes the

> > logic of terms much cleaner.

>

> Yes, but apparently it does so by breaking past usage of {lo}.

For example? Only if you were relying on the implicit quantifier

to transmit something significant is the usage broken. I don't

see many examples of that.

> Sorry, but this strikes me as one of the areas where {lo} is _supposed_ to be

> different from {le} and {la} - it can refer to different things at different

> times. Making {lo} generic is great. But if it's generic, it has to be able

> to

> mean {su'o} in some contexts, and {su'o} is not a constant.

E>{lo} allows that su'o is true, but it does not ever impose it.

Context may strongly suggest {su'o}, and the semantics of the

predicate may strongly require it (how can you see something

if you don't see at least one instance?) but the quantifier

is not part of the bare {lo} claim.

> Sorry if I'm getting pissy. You were _so_ close to an entirely agreeable

> proposal about {lo} that handled all past usage, but now constants are

> smeared

> all over it.

F>If you are not much concerned with the logic part of a

Lojban sentence you can disregard all talk about constants.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 16:50 GMT

On Wed, Jun 09, 2004 at 07:10:19AM -0700, [email protected] wrote:

> Constants

> Some comments on constant terms.

>

> IF X and Z are constant terms, then the following equivalences

> always hold:

>

> X na broda == X naku broda == naku X broda == X broda naku

> X broda Z == Z se broda X

> X broda Z ije X brode Z == X boi Z broda gi'e brode

> X broda lo nu Z brode ije ri brodi == X broda lo nu Z brode ije Z brodi

>

> None of the above work in general with terms that carry an

> outer quantifier.

...

Do I take you to mean that you think xorlo would make {lo broda}

into a "constant term"?

Please note that this *drastically* changes the meaning of some

lojban sentences. As well as completely breaking negation with

{naku}.

For example:

lo curnu cu na nenri le plise

old: == naku lo curnu cu nenri le plise ("It is false that a worm is in the apple")

Which is same as the strong claim that

== ro curnu naku nenri le plise ("All worms are not in the apple")

xorlo: lo curnu cu na nenri le plise

== lo curnu cu naku nenri le plise

xorxes can provide a gloss. "worm not in apple", by his earlier thoughts

Which no longer implies that every worm is not in the apple. In

fact, it would allow both

lo curnu cu na nenri le plise

and

lo curnu cu nenri le plise

to be simultaneously true.

Which completely breaks the consistency of the language, because

putting {na} before the selbri is the same as putting {naku} before

the whole deal, and presumably any true sentence negated by {naku}

should be false....

{naku lo curnu cu nenri} cannot equal {lo curnu naku nenri}.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxy5bDrrilS51AZ8RAlMXAJ0fYl5OvNw0g1IuYkZwimA1Ut7yrwCePx0W

V+Mbi8Gu62oZJJzcH3MoiWI=

=DEWY

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 16:50 GMT

Jorge Llamb?as scripsit:

> {ci gerku cu batci le pa bongu}.

Which is to say, "ci [lo pa] gerku".

--

Here lies the Christian, John Cowan

judge, and poet Peter, http://www.reutershealth.com

Who broke the laws of God http://www.ccil.org/~cowan

and man and metre. [email protected]



Posted by Anonymous on Wed 09 of June, 2004 16:50 GMT

Jordan:

> ...

> > > lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> > > dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

>

> The old {lo} does impose that it is a real, existing event, which

> occurs for at least one fox, who saw at least one tree, etc.

>

> Correct me if I'm wrong, but xorlo doesn't require that the fox

> actually exists. It could work just as well for me to say:

>

> lo pavyseljirna noi xagji ku'o ca lenu viska loi srasu noi

> cpana lo rajma'a cu djica lenu cpacu ra gi'e naka'e cpacu

Since we are talking about a fable here, I don't think it being

about lorxu or pavyseljirna makes any difference. If I remember

correctly the lorxu later has a conversation with the cipnrkorvo.

> (Of course the above could work in old-lo, but only in a work of

> fiction---no one will believe the speaker is talking about reality)

No one is supposed to believe that.

> > > lojbo: lo since cu nenri lo sipna kumfa

>

> What it means is

> Exx) . Ey(sipna_kumfa(y) . nenri(x, y?)

That's what it means with {su'o lo}, right.

With bare {lo}, I propose it should rather mean:

nenri(SINCE, SIPNA KUMFA)

No quantification implied.

> If you want that, why not say {since nenri kumfa}?

That's much more vague. It doesn't say that SINCE is in the x1

of nenri and KUMFA in the x2.

> You didn't adress my point about this killing the only well-defined

> article in the language. I think BPFK article changes should really

> go towards further formalization, rather than away from the small

> amount we currently have.

I think defining every not explicitly quantified term as not

quantified is a big step in that direction.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

G> Well, xorxes occasionally talks as though it did not, but it has to for what he wants it to do (but which it can't do anyhow as presently constituted).

H> Presumably because that means "Lo, the snake-inside type of room," which is not the utterance anyone around was trying to make (though I agree it is a handier way of doing things could we get the semantics and grammar to cooperate).

Jordan DeLong wrote:

...

> > Nick's aesop stuff:

> > lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> > dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

> > xorlo: Foxes are hungry and when they see some grapes

> > dangling from trees want to take them but can't.

> > old-lo: A fox, who was hungry, upon seeing some grapes in a tree

> > wanted to take them, but couldn't.

>

> Why do you translate the second as a particular event and the first one

> as a general event, given that there is no indication either way in

> the Lojban? old-lo could just as well be: "some foxes are hungry

> and when they see some grapes dangling from some trees want to

> take them but can't." The proposed {lo} does not impose a generic

> reading to the whole sentence, it is just not explicit as to instances.

> All it says is: "Fox, hungry, when sees that grape hangs from tree, wants

> get it but can't get". How you choose to fill the gaps must come from

> context.

The old {lo} does impose that it is a real, existing event, which

occurs for at least one fox, who saw at least one tree, etc.

G>Correct me if I'm wrong, but xorlo doesn't require that the fox

actually exists. It could work just as well for me to say:

lo pavyseljirna noi xagji ku'o ca lenu viska loi srasu noi

cpana lo rajma'a cu djica lenu cpacu ra gi'e naka'e cpacu

(Of course the above could work in old-lo, but only in a work of

fiction---no one will believe the speaker is talking about reality)

> > From my translation of Ambrose Bierce's "The Man and the Snake":

> >

> > lojbo: lo since cu nenri lo sipna kumfa

> > xorlo: Snakes are inside of bedrooms.

> > old-lo: There is a snake for whom there is a bedroom containing him.

>

> or: At least one snake is inside of at least one room.

Same thing. My gloss was intended to emphasize the logical structure

of it.

What it means is

Exx) . Ey(sipna_kumfa(y) . nenri(x, y?)

> With the proposed lo, I would simply say "snake inside room". How you

> turn that into more fluent English will depend on the context. {lo}

> adds nothing.

H>If you want that, why not say {since nenri kumfa}?

You didn't adress my point about this killing the only well-defined

article in the language. I think BPFK article changes should really

go towards further formalization, rather than away from the small

amount we currently have.

--

Jordan DeLong

[email protected]



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

Lord, I hope not: singletons don't bite bones (I am not sure about triads, but probably not). I suspect this is just a terminological problems, though working out the logic is a bitch — because of the quantifiers, as usual.

[email protected] wrote:Jorge Llamb?as scripsit:

> {ci gerku cu batci le pa bongu}.

Which is to say, "ci [lo pa] gerku".

--

Here lies the Christian, John Cowan

judge, and poet Peter, http://www.reutershealth.com

Who broke the laws of God http://www.ccil.org/~cowan

and man and metre. [email protected]



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

H> That is exactly what the quantifier is.

I> Except that it doesn't work, and it leaves them without any reliable logic at all.

Jorge Llambías wrote:

Jordan:

> ...

> > > lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> > > dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

>

> The old {lo} does impose that it is a real, existing event, which

> occurs for at least one fox, who saw at least one tree, etc.

>

> Correct me if I'm wrong, but xorlo doesn't require that the fox

> actually exists. It could work just as well for me to say:

>

> lo pavyseljirna noi xagji ku'o ca lenu viska loi srasu noi

> cpana lo rajma'a cu djica lenu cpacu ra gi'e naka'e cpacu

Since we are talking about a fable here, I don't think it being

about lorxu or pavyseljirna makes any difference. If I remember

correctly the lorxu later has a conversation with the cipnrkorvo.

> (Of course the above could work in old-lo, but only in a work of

> fiction---no one will believe the speaker is talking about reality)

No one is supposed to believe that.

> > > lojbo: lo since cu nenri lo sipna kumfa

>

> What it means is

> Exx) . Ey(sipna_kumfa(y) . nenri(x, y?)

That's what it means with {su'o lo}, right.

With bare {lo}, I propose it should rather mean:

nenri(SINCE, SIPNA KUMFA)

H>No quantification implied.

> If you want that, why not say {since nenri kumfa}?

That's much more vague. It doesn't say that SINCE is in the x1

of nenri and KUMFA in the x2.

> You didn't adress my point about this killing the only well-defined

> article in the language. I think BPFK article changes should really

> go towards further formalization, rather than away from the small

> amount we currently have.

I>I think defining every not explicitly quantified term as not

quantified is a big step in that direction.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

Jordan:

> Do I take you to mean that you think xorlo would make {lo broda}

> into a "constant term"?

No quantifier would ever be implicitly imposed, correct.

> Please note that this *drastically* changes the meaning of some

> lojban sentences.

Not any of the examples you have presented.

> {naku lo curnu cu nenri} cannot equal {lo curnu naku nenri}.

By that you mean that {naku su'o lo curnu cu nenri}

cannot equal {su'o lo curnu naku nenri}, something that

I'm not disputing.

I'm saying that without an explicit su'o, it is not a quantified

claim. It just means:

~ nenri(CURNU, )

and it is left to the context to fill the gaps. When? Where?

For every instance? Inside of what? etc.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

On Wed, Jun 09, 2004 at 08:37:56AM -0700, Jorge Llambías wrote:

> Jordan:

> > Do I take you to mean that you think xorlo would make {lo broda}

> > into a "constant term"?

>

> No quantifier would ever be implicitly imposed, correct.

>

> > Please note that this *drastically* changes the meaning of some

> > lojban sentences.

>

> Not any of the examples you have presented.

Currently {lo curnu cu na nenri le plise} means that *all* worms

in the entire world (if there are any worms) are not in the apple.

This meaning is not preserved if the naku can't move and make {su'o

lo curnu} into an assertion about {ro curnu}.

> > {naku lo curnu cu nenri} cannot equal {lo curnu naku nenri}.

>

> By that you mean that {naku su'o lo curnu cu nenri}

> cannot equal {su'o lo curnu naku nenri}, something that

> I'm not disputing.

...

No, I mean under any meaning of {lo}. If you allow {naku} to move

like that, you change the language so that prefixing a true assertion

with {naku} no longer produces a false assertion. In lojban I

believe this currently works for all sentences which have a

truth-value.

Do you agree that under xorlo {lo curnu cu nenri le plise} does not

imply that {naku lo curnu cu nenri le plise} is false?

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxzauDrrilS51AZ8RAlbkAJ9CsghnTx91Fy/sdV/gKIQQdZ4fvACfedIb

mpEuRw0AQo+Om1WqKJy1UA4=

=PIey

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

On Wed, Jun 09, 2004 at 08:23:03AM -0700, Jorge Llambías wrote:

> Jordan:

> > ...

> > > > lojbo: .i lo lorxu noi xagji ku'o ca lenu viska loi vanjba noi

> > > > dandu lo tricu cu djica lenu cpacu ra gi'e naka'e cpacu

> >

> > The old {lo} does impose that it is a real, existing event, which

> > occurs for at least one fox, who saw at least one tree, etc.

> >

> > Correct me if I'm wrong, but xorlo doesn't require that the fox

> > actually exists. It could work just as well for me to say:

> >

> > lo pavyseljirna noi xagji ku'o ca lenu viska loi srasu noi

> > cpana lo rajma'a cu djica lenu cpacu ra gi'e naka'e cpacu

>

> Since we are talking about a fable here, I don't think it being

> about lorxu or pavyseljirna makes any difference. If I remember

> correctly the lorxu later has a conversation with the cipnrkorvo.

It's just an example sentence; that it's from a fable is a nonissue.

An equivalent sentence for our purposes can be constructed which

isn't from a work of fiction.

But in the context of the fable, we're supposed to believe it is

true. The fable creates an imaginary world---in that world there

really is at least one fox who saw a tree and so on. The {lo}

asserts its existence---it is not a generic statement about how

foxes behave when they see berries in a tree.

...

> > > > lojbo: lo since cu nenri lo sipna kumfa

> >

> > What it means is

> > Exx) . Ey(sipna_kumfa(y) . nenri(x, y?)

>

> That's what it means with {su'o lo}, right.

> With bare {lo}, I propose it should rather mean:

>

> nenri(SINCE, SIPNA KUMFA)

>

> No quantification implied.

Right. Which is a very big change to the meaning of the sentence,

as well as a reduction in formalism.

...

> > You didn't adress my point about this killing the only well-defined

> > article in the language. I think BPFK article changes should really

> > go towards further formalization, rather than away from the small

> > amount we currently have.

>

> I think defining every not explicitly quantified term as not

> quantified is a big step in that direction.

OTOH, I think all terms (excepting non-logical ones like ma) should

be definable in terms of quantified variables.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAxzhpDrrilS51AZ8RAk20AJ9U6KyBOQoKD7SK5nR24Zg5xKFM5wCfS7FN

II7JMzZ2lODul+YKiQbTSSc=

=Uqjw

---END PGP SIGNATURE---



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

Jordan:

> > > Please note that this *drastically* changes the meaning of some

> > > lojban sentences.

> >

> > Not any of the examples you have presented.

> Currently {lo curnu cu na nenri le plise} means that *all* worms

> in the entire world (if there are any worms) are not in the apple.

> This meaning is not preserved if the naku can't move and make {su'o

> lo curnu} into an assertion about {ro curnu}.

Sorry, I thought you meant usage sentences. Usage sentences usually

have a context so that such absolute terms hardly apply.

> Do you agree that under xorlo {lo curnu cu nenri le plise} does not

> imply that {naku lo curnu cu nenri le plise} is false?

Give me context!

{mi nenri le mi zdani} does not necessarily imply that

{naku mi nenri le mi zdani} is false, I need to know if you want

to keep every other possible circumstance constant or not.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

Rob Speer wrote:

>On Wed, Jun 09, 2004 at 07:10:19AM -0700, [email protected] wrote:

>

>

>>With the proposal, every Lojban sumti with no explicit

>>outer quantifier is a constant (i.e. it carries no implicit quantifier)

>>and always permits the above transformations. (Even terms

>>like da behave that way when they are within the scope of

>>their quantifier and thus carry no outer quantifier. They

>>can't be manipulated like that when they do carry an outer

>>quantifier, you must first send the quantifier to a prenex.)

>>

>>With CLL-defaults, {le broda} and {la broda} are pseudo-constants,

>>they carry a quantifier but they usually have a single referent, so

>>the transformations "usually" work. The proposal, by requiring

>>explicit quantifiers when quantification is intended, makes the

>>logic of terms much cleaner.

>>

>>

>

>Yes, but apparently it does so by breaking past usage of {lo}.

>

>Sorry, but this strikes me as one of the areas where {lo} is _supposed_ to be

>different from {le} and {la} - it can refer to different things at different

>times. Making {lo} generic is great. But if it's generic, it has to be able to

>mean {su'o} in some contexts, and {su'o} is not a constant.

>

>Sorry if I'm getting pissy. You were _so_ close to an entirely agreeable

>proposal about {lo} that handled all past usage, but now constants are smeared

>all over it.

>

>

In order to even mention past usage as a factor, you have to show that

the past usage is meaningful: that it was used in a consistent way by

people who understood how they were using it, and with all

contradictions recognized by consensus as *mistakes*. Otherwise, past

usage breaks past usage, and you are restricting us with the impossible

task of upholding contradiction.

The only way this new lo was able to get out of committee — having been

kicked around for easily a year on jboske, the record of which you all

are invited to study at http://groups.yahoo.com/group/jboske/ — is the

fact that su'oda is available for your logical precision pleasure.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

Jorge Llamb?as scripsit:

> > Do you agree that under xorlo {lo curnu cu nenri le plise} does not

> > imply that {naku lo curnu cu nenri le plise} is false?

>

> Give me context!

It doesn't matter what the context may be as long as it's fixed.

--

John Cowan [email protected] http://www.ccil.org/~cowan

Does anybody want any flotsam? / I've gotsam.

Does anybody want any jetsam? / I can getsam.

--Ogden Nash, No Doctors Today, Thank You



Posted by Anonymous on Wed 09 of June, 2004 16:51 GMT

On Wed, Jun 09, 2004 at 11:11:26AM -0500, Jordan DeLong wrote:

> No, I mean under any meaning of {lo}. If you allow {naku} to move

> like that, you change the language so that prefixing a true assertion

> with {naku} no longer produces a false assertion. In lojban I believe

> this currently works for all sentences which have a truth-value.

You're joking, right?

mi vofli

do na ka'e vofli

mi pu vofli se pi'o le vinji

All of these sentences are true.

Remember, neither the tense nor CAhA status are actually part of the

sentence unless specified.

I don't see a substantial difference between that an the new lo stuff.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 17:17 GMT

The new notation here (OK, so I've seen it once before today) suggests that {lo broda} is now being taken as a species or a property or a set or some such abstract thing. This makes perfectly good sense and fits more of xorxes tales than Mr. Rabbit, a "concrete object." But it does raise another set of problems: how to tell the difference between what is said about the abstract in the web of abstractions from what is said of it in the web of facts, for example, how to tell whether nenri(curnu, plise) merely says that worms are the sort of things that might be inside apples or says that worms actually are inside apples from time to time and place to place. Tenseless Lojban sentences have the same problem (officially) and it is resolved (offically and actually) by context. But, just in case, we need also an offical way (as we do with vague tenses like {su'oroi}) to disambiguate without losing too much of the original intent. Such ways exist, and, if we decide to use {lo} in this

way, will have to explored. While the basic sentences are still the same, the disambiguations are mildly hairy.

Jorge Llambías wrote:

Jordan:

> Do I take you to mean that you think xorlo would make {lo broda}

> into a "constant term"?

No quantifier would ever be implicitly imposed, correct.

> Please note that this *drastically* changes the meaning of some

> lojban sentences.

Not any of the examples you have presented.

> {naku lo curnu cu nenri} cannot equal {lo curnu naku nenri}.

By that you mean that {naku su'o lo curnu cu nenri}

cannot equal {su'o lo curnu naku nenri}, something that

I'm not disputing.

I'm saying that without an explicit su'o, it is not a quantified

claim. It just means:

~ nenri(CURNU, )

and it is left to the context to fill the gaps. When? Where?

For every instance? Inside of what? etc.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 17:17 GMT

On Wed, Jun 09, 2004 at 12:19:29PM -0400, [email protected]

wrote:

> Jorge Llamb?as scripsit:

>

> > > Do you agree that under xorlo {lo curnu cu nenri le plise} does

> > > not imply that {naku lo curnu cu nenri le plise} is false?

> >

> > Give me context!

>

> It doesn't matter what the context may be as long as it's fixed.

That's not his point; you snipped the important part:

{mi nenri le mi zdani} does not necessarily imply that {naku mi

nenri le mi zdani} is false, I need to know if you want to keep

every other possible circumstance constant or not.

Note that that sentence does not use "lo".

It has *never* been the case that a sentence without *everything*

specified (tense & CAhA, at a bare minimum) always asserted that its

negation was untrue.

mi ca nenri le mi zdani == mi nenri le mi zdani

mi pu na nenri le mi zdani == mi na nenri le mi zdani

If you think "mi nenri le mi zdani" always implies that "mi na nenri le

mi zdani" is false, without context, you are living in a fantasy world.

Nevermind CAhA, which just makes things worse, or the tricks I could

pull with BAI.

I've never used "lo" in this post, but I'm able to produce the exact

same problems that Jordan (and, apparently, you) are complaining about.

If you want these problems to go away (hah!) it's going to take a hell

of a lot more than changing xorlo.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 17:17 GMT

Well, suppose that we intend to keep all the preverb the same in both cases, whatever it unspecifiedly is. That is, suppose that the {naku} sentence is meant to be exactly the denial of the unnegated form. This is, after all, what the question was.

Jorge Llambías wrote:

Jordan:

> > > Please note that this *drastically* changes the meaning of some

> > > lojban sentences.

> >

> > Not any of the examples you have presented.

> Currently {lo curnu cu na nenri le plise} means that *all* worms

> in the entire world (if there are any worms) are not in the apple.

> This meaning is not preserved if the naku can't move and make {su'o

> lo curnu} into an assertion about {ro curnu}.

Sorry, I thought you meant usage sentences. Usage sentences usually

have a context so that such absolute terms hardly apply.

> Do you agree that under xorlo {lo curnu cu nenri le plise} does not

> imply that {naku lo curnu cu nenri le plise} is false?

Give me context!

{mi nenri le mi zdani} does not necessarily imply that

{naku mi nenri le mi zdani} is false, I need to know if you want

to keep every other possible circumstance constant or not.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 17:17 GMT

"Past usage" is a generic term — it does not require that *all* past usage be covered, nor even that most of it be. In the case of Lojban, we actually have a normative description of what past usage *should* be, against which we may discover that much actual past usage fails. But what remains, even if it is only one passage, has a special claim on being supported by any change. Actually, it seems that most past usage is pretty much OK, although the pragmatics of the situations have not been explored or exploited. the criticisms have generally been from that part of the issue — and have generally been as misleading andd misled as the examples criticized.

Even if we have {su'o} for greater precision, it still remains a question whether the new {lo} does anything useful, let alone whether it does what it is claimed to do. And whether ti does it better than old {lo}.

xod wrote:

In order to even mention past usage as a factor, you have to show that

the past usage is meaningful: that it was used in a consistent way by

people who understood how they were using it, and with all

contradictions recognized by consensus as *mistakes*. Otherwise, past

usage breaks past usage, and you are restricting us with the impossible

task of upholding contradiction.

The only way this new lo was able to get out of committee — having been

kicked around for easily a year on jboske, the record of which you all

are invited to study at http://groups.yahoo.com/group/jboske/ — is the

fact that su'oda is available for your logical precision pleasure.



Posted by Anonymous on Wed 09 of June, 2004 17:17 GMT

On Wed, Jun 09, 2004 at 08:04:21AM -0700, Jorge Llamb?as wrote:

> For example? Only if you were relying on the implicit quantifier

> to transmit something significant is the usage broken. I don't

> see many examples of that.

I have no problem with the implicit quantifiers being removed; I think it's a

good idea.

But the claim that it sounds like you're making is that {lo jvugi} *always*

refers to jvugis in general. This may not be what you mean, but it sounds like

what you're saying.

The way I read the proposal when I liked it, {lo jvugi} could mean any of

  • one jvugi
  • many jvugis
  • jvugis in general
  • intensional jvugis

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

Rob:

> The way I read the proposal when I liked it, {lo jvugi} could mean any of

> * one jvugi

> * many jvugis

> * jvugis in general

> * intensional jvugis

The way I understand it, lo jvugi does not mean any of them,

but can cover cases where any of them would apply. When

you use {lo jvugi} you are not making the claim that one jvugi

does this or that, nor that jvugis in general do this

or that. All you are doing is making the x1 of jvugi coincide

with whatever slot you fill and leaving everything else to

context (or other precision words, like possibly quantifiers

over instances).

Outside of all context, sentences with {lo jvugi} are easiest

to interpret as claims about jvugis in general, but that's

just the way out of context sentences work. If you have a

jvugi right in front of view and say {lo jvugi ca ca'o clani}

you may be just reflecting what you see: jvugi is being long.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

On Wed, Jun 09, 2004 at 10:28:28AM -0700, Jorge Llamb?as wrote:

>

> Rob:

> > The way I read the proposal when I liked it, {lo jvugi} could mean

> > any of

> > * one jvugi

> > * many jvugis

> > * jvugis in general

> > * intensional jvugis

>

> The way I understand it, lo jvugi does not mean any of them,

> but can cover cases where any of them would apply.

Dude, you write kick-ass proposals, but you don't explain them so good.

Rob: Yes, you are correct.

:-)

> Outside of all context, sentences with {lo jvugi} are easiest

> to interpret as claims about jvugis in general, but that's

> just the way out of context sentences work.

I think that sentence is important to keep in mind.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

Jordan:

> OTOH, I think all terms (excepting non-logical ones like ma) should

> be definable in terms of quantified variables.

Consider the term {zo lo}.

zo lo cmavo

"lo" is a cmavo

mi ciska reno zo lo le papri

I write 20 "lo"s on the page.

Would it even cross you mind to think of instances of zo lo

in the first sentence? The second sentence clearly deals with

instances, the first one doesn't.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

John E Clifford scripsit:

> > > {ci gerku cu batci le pa bongu}.

> >

> > Which is to say, "ci [lo pa] gerku".

> >

> Lord, I hope not: singletons don't bite bones (I am not sure about

> triads, but probably not). I suspect this is just a terminological

> problems, though working out the logic is a bitch — because of the

> quantifiers, as usual.

When there is an outer quantifier to lo, but no inner quantifier, it

retains its meaning of quantifying over individuals, for the sake of

backward compatibility. This is a wart, but a necessary one.

--

We call nothing profound [email protected]

that is not wittily expressed. John Cowan

--Northrop Frye (improved) http://www.reutershealth.com



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

pc:

> for

> example, how to tell whether nenri(curnu, plise) merely says that worms are

> the sort of things that might be inside apples or says that worms actually

> are inside apples from time to time and place to place.

{ka'e}, {ca'a}, {su'oroi}, {so'iroi}, {fe'e su'oroi}, {fe'eso'iroi}, etc.

> Tenseless Lojban

> sentences have the same problem (officially) and it is resolved (offically

> and actually) by context.

Right.

> But, just in case, we need also an offical way (as

> we do with vague tenses like {su'oroi}) to disambiguate without losing too

> much of the original intent. Such ways exist, and, if we decide to use {lo}

> in this

> way, will have to explored. While the basic sentences are still the same,

> the disambiguations are mildly hairy.

We have, of course, explicit quantifiers to use when needed.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 18:04 GMT

Robin Lee Powell scripsit:

> It has *never* been the case that a sentence without *everything*

> specified (tense & CAhA, at a bare minimum) always asserted that its

> negation was untrue.

>

> mi ca nenri le mi zdani == mi nenri le mi zdani

>

> mi pu na nenri le mi zdani == mi na nenri le mi zdani

>

> If you think "mi nenri le mi zdani" always implies that "mi na nenri le

> mi zdani" is false, without context, you are living in a fantasy world.

I see your meta-argument and raise you one meta.

No sentence has any meaning at all without a specification of context,

because some things must always be left to the context. There are after

all an infinite amount of places that could be attached to a bridi --

one for each place of each possible selbri, using fi'o.

Now, given a particular context, the bridi "naku broda" and "broda" are

contradictories: if one is true, the other is false, for any binding of

"broda" whatsoever. Any argument that purports to upset that is bogus,

bogus, and more bogus. By the same token, in two different contexts, no

statements can be made about the relationship between the truth values of

"broda" and "naku broda".

I agree that this has nothing to do with "lo" specifically.

--

"Well, I'm back." --Sam John Cowan



Posted by Anonymous on Wed 09 of June, 2004 18:45 GMT

On Wed, Jun 09, 2004 at 01:54:51PM -0400, [email protected] wrote:

> Now, given a particular context, the bridi "naku broda" and "broda"

> are contradictories: if one is true, the other is false, for any

> binding of "broda" whatsoever. Any argument that purports to upset

> that is bogus, bogus, and more bogus.

Well, there's "le'e", but I suppose the person speaking counts as part

of the context.

What about the mass gadri? I thought it was part of the lore of masses

that they had all the features of their members, such that the mass of

all lions is both male and female?

> By the same token, in two different contexts, no statements can be

> made about the relationship between the truth values of "broda" and

> "naku broda".

>

> I agree that this has nothing to do with "lo" specifically.

Good.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

jcowan:

> Now, given a particular context, the bridi "naku broda" and "broda" are

> contradictories: if one is true, the other is false, for any binding of

> "broda" whatsoever. Any argument that purports to upset that is bogus,

> bogus, and more bogus.

Agreed. Sometimes though, given a context it is hard to determine

what exactly the speaker means with a vague sentence. Suppose we

are in front of a wall with white and yellow spots. Someone asks

{xu le bitmu cu pelxu}.

The natural answer is neither {ja'a go'i} nor {na go'i},

but something more explicative. Either answer could be defended

depending on your suppositions.

Similarly, if there are some jvugis inside and others outside

the box, and someone asks {xu lo jvugi cu nenri}, "jvugi inside?"

the most natural answer is neither {ja'a go'i} nor {na go'i} but

something more explicative (like {su'oboi jy go'i}). Both answers

can be defended depending on your suppositions.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

On Wed, Jun 09, 2004 at 08:37:56AM -0700, Jorge Llamb?as wrote:

> By that you mean that {naku su'o lo curnu cu nenri}

> cannot equal {su'o lo curnu naku nenri}, something that

> I'm not disputing.

But you do say that {naku lo curnu cu nenri} equals {lo curnu naku nenri},

and you say that {lo} can mean {su'o lo}, so there's a problem somewhere.

What advantage does negation transparency give us that it's worth all this

metaphysical debate to get?

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

Robin Lee Powell scripsit:

> What about the mass gadri? I thought it was part of the lore of masses

> that they had all the features of their members, such that the mass of

> all lions is both male and female?

Very true. So "loi cinfo cu nakni" is true (and "loi cinfo cu na nakni"

is false) and "loi cinfo cu fetsi" is true (and "loi cinfo cu na fetsi"

is false). This is not a *contradiction*, just surprising.

--

John Cowan [email protected] www.ccil.org/~cowan

Female celebrity stalker, on a hot morning in Cairo:

"Imagine, Colonel Lawrence, ninety-two already!"

El Auruns's reply: "Many happy returns of the day!"



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

On Wednesday 09 June 2004 15:40, [email protected] wrote:

> Very true. So "loi cinfo cu nakni" is true (and "loi cinfo cu na nakni"

> is false) and "loi cinfo cu fetsi" is true (and "loi cinfo cu na fetsi"

> is false). This is not a *contradiction*, just surprising.

Also {loi cinfo naku nakni} is true, as is {loi cinfo naku fetsi}. However,

{loi figre naku fetsi} is false: ro figre bezu'i cu fetsi po'o (to se grute

le se citka toi) gi'onai fetsi joibo nakni (to kanbyfigre toi). iji'a da poi

se skargolu zo'u lo nakni cu binxo lo fetsi, and I'll let you figure out what

to say about loi skargolu be da.

mu'omi'e pier.

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

On Wed, Jun 09, 2004 at 03:56:19PM -0400, Pierre Abbat wrote:

> On Wednesday 09 June 2004 15:40, [email protected] wrote:

> > Very true. So "loi cinfo cu nakni" is true (and "loi cinfo cu na

> > nakni" is false) and "loi cinfo cu fetsi" is true (and "loi cinfo cu

> > na fetsi" is false). This is not a *contradiction*, just

> > surprising.

>

> Also {loi cinfo naku nakni} is true, as is {loi cinfo naku fetsi}.

Help the poor confused person, please.

What's the difference between

"loi cinfo naku nakni" and "loi cinfo na nakni"?

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

Robin Lee Powell scripsit:

> What's the difference between

>

> "loi cinfo naku nakni" and "loi cinfo na nakni"?

In XS (Xorxes's Solution), nothing, as best I understand.

In CLL, the former meant "The mass-of-all-lions is not wholly male",

and the latter meant "The mass-of-all-lions is not male in any part".

That was because "loi" = "pisu'o loi", so the pisu'o changes to piro as

the negation boundary passes over it. (In XS, if you want "pisu'o loi",

you say so.)

The two can be paraphrased more narrowly as "For some part of the

total-lion-mass, it is false that it is male" and "It is false that some

part of the total-lion-mass is male."

In any case, the first is true and the second is false.

--

"And it was said that ever after, if any John Cowan

man looked in that Stone, unless he had a [email protected]

great strength of will to turn it to other www.ccil.org/~cowan

purpose, he saw only two aged hands withering www.reutershealth.com

in flame." --"The Pyre of Denethor"



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

On Wednesday 09 June 2004 16:13, Robin Lee Powell wrote:

> Help the poor confused person, please.

>

> What's the difference between

>

> "loi cinfo naku nakni" and "loi cinfo na nakni"?

{loi cinfo na nakni} = {naku loi cinfo cu nakni}. Since many of the lions are

male, the mass is male, so that's false.

{loi cinfo naku nakni}. Since many of the lions are non-male, the mass is

non-male, so that's true.

But among figs, all are female, but only some are male. So {loi figre naku

fetsi} is false.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Wed 09 of June, 2004 21:46 GMT

On Wed, Jun 09, 2004 at 12:19:29PM -0400, [email protected] wrote:

> Jorge Llamb?as scripsit:

> Jordan:

> > > Do you agree that under xorlo {lo curnu cu nenri le plise} does

> > > not imply that {naku lo curnu cu nenri le plise} is false?

> >

> > Give me context!

>

> It doesn't matter what the context may be as long as it's fixed.

All of my response deals only with xorlo.

For any *fixed* context, including the binding of the constant (which is

only in the speaker's head), "lo curnu cu nenri le plise" implies that

"naku lo curnu cu nenri le plise" is false (assuming I understand naku,

which is false).

However, there is nothing the requires the speaker to maintain the

binding of the constant from one statement to the next, and that's where

the problem is.

There is nothing wrong with the following interchange:

A: lo curnu cu nenri le plise

B: .i ma curnu

A: ro curnu poi nenri le plise .iji'a naku lo curnu cu nenri le plise

B: .i .oisai ma curnu

A: ro curnu poi na nenri le plise

My question is, is "lo curnu cu nenri le plise" false if there is, in

fact, no worm in the apple? I believe that it is false in that case,

but I'd like others' opinions. Assuming I'm right, the above exchange

is only sensible if there is at least one worm in the apple.

Now, if it was "lo curnu cu nenri *lo* plise", that would be different.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 11:52:25AM -0400, Rob Speer wrote:

> On Wed, Jun 09, 2004 at 08:37:56AM -0700, Jorge Llamb?as wrote:

> > By that you mean that {naku su'o lo curnu cu nenri} cannot equal

> > {su'o lo curnu naku nenri}, something that I'm not disputing.

>

> But you do say that {naku lo curnu cu nenri} equals {lo curnu naku

> nenri}, and you say that {lo} can mean {su'o lo}, so there's a problem

> somewhere.

"lo" can mean "su'o lo" in the sense that "lo broda" can mean "One or

more broda", but it is not logically quantified at any point.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

Robin Lee Powell scripsit:

> For any *fixed* context, including the binding of the constant (which is

> only in the speaker's head), "lo curnu cu nenri le plise" implies that

> "naku lo curnu cu nenri le plise" is false (assuming I understand naku,

> which is false).

Naku at the beginning is easy: it's exactly "It is false that ...".

> However, there is nothing the requires the speaker to maintain the

> binding of the constant from one statement to the next, and that's where

> the problem is.

Granted, but if the speaker varies things too wildly, the listener will

get confused, and Lojban is intended to be listener-centric: if the

listener doesn't get it, the speaker has a (Gricean) obligation to

amplify.

> A: lo curnu cu nenri le plise

> B: .i ma curnu

> A: ro curnu poi nenri le plise .iji'a naku lo curnu cu nenri le plise

> B: .i .oisai ma curnu

> A: ro curnu poi na nenri le plise

All serene, provided some but not all worms are in the apple. One can

construct similar dialogues without using anything like lo, however:

A: rants about Socrates

B: Who is Socrates?

A: train of thought derailed Why, Socrates is the person I call "Socrates",

to be sure.

B: Yes, but whom do you call "Socrates"?

A: Why, Socrates, of course. Do pay attention. resumes rant

> My question is, is "lo curnu cu nenri le plise" false if there is, in

> fact, no worm in the apple? I believe that it is false in that case,

Subject to reasonable assumptions (which are themselves part of the

context), such as that the tense is "caca'a" and that there is no

implicit BAI or FIhO hanging about, and so on: then yes.

> Assuming I'm right, the above exchange

> is only sensible if there is at least one worm in the apple.

I think so too. The snag comes in when we say "lo -unicorn cu blabi", there

being no unicorns. I think we either assume we're in a fantasy world, or

we have to reword this as talk of concepts rather than things.

> Now, if it was "lo curnu cu nenri *lo* plise", that would be different.

This is falsified if there are no wormy apples, IMHO.

--

John Cowan http://www.ccil.org/~cowan

Raffiniert ist der Herrgott, aber boshaft ist er nicht.

--Albert Einstein



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 04:58:52PM -0400, [email protected] wrote:

> Robin Lee Powell scripsit:

> > My question is, is "lo curnu cu nenri le plise" false if there is,

> > in fact, no worm in the apple? I believe that it is false in that

> > case,

>

> Subject to reasonable assumptions (which are themselves part of the

> context), such as that the tense is "caca'a" and that there is no

> implicit BAI or FIhO hanging about, and so on: then yes.

>

> > Assuming I'm right, the above exchange is only sensible if there is

> > at least one worm in the apple.

>

> I think so too. The snag comes in when we say "lo -unicorn cu blabi",

> there being no unicorns. I think we either assume we're in a fantasy

> world, or we have to reword this as talk of concepts rather than

> things.

Actually, that causes no problems WRT the nenri example. If one

translates "lo pavyseljirna cu blabi" as "The idea of unicorns is white"

(as you are welcome to do; xorlo is empty after all), one could then

complain that "lo curnu cu nenri le plise" should be translated as "The

idea of a worm is in this apple", to which I would respond that, since

you are talking about a particular apple, it's contents are precisely

determinable and all of them can be compared to the idea of a worm for a

match if you wish.

Then I would slap one for being annoying.

> > Now, if it was "lo curnu cu nenri *lo* plise", that would be

> > different.

>

> This is falsified if there are no wormy apples, IMHO.

Agreed, as long as there have *never* been any wormy apples, even in

someone's head:

ca lo purlamcte mi senva la'e di'e .i lo curnu cu pu nenri lo plise

The xorlo mantra: "xorlo adds nothing".

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 10:32:28AM -0700, Robin Lee Powell wrote:

> On Wed, Jun 09, 2004 at 10:28:28AM -0700, Jorge Llamb?as wrote:

> >

> > Rob:

> > > The way I read the proposal when I liked it, {lo jvugi} could mean

> > > any of

> > > * one jvugi

> > > * many jvugis

> > > * jvugis in general

> > > * intensional jvugis

> >

> > The way I understand it, lo jvugi does not mean any of them,

> > but can cover cases where any of them would apply.

>

> Dude, you write kick-ass proposals, but you don't explain them so good.

>

> Rob: Yes, you are correct.

You see, I don't want {lo} to mean some weird concept that is supposed to

be all of those at the same time. You're having a difficult time convincing

most people that such a concept makes any sense at all. And though PC can be a

pain, he's also a logician, and I will trust him when he says your overarching

concept makes no sense.

I want {lo} to literally mean {su'o} some of the time, because that's how you

encompass past usage.

So which does it mean?

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 12:09:58PM -0400, xod wrote:

> In order to even mention past usage as a factor, you have to show that

> the past usage is meaningful: that it was used in a consistent way by

> people who understood how they were using it, and with all

> contradictions recognized by consensus as *mistakes*. Otherwise, past

> usage breaks past usage, and you are restricting us with the impossible

> task of upholding contradiction.

Please clarify: are you saying that {lo} has never been used correctly to mean

{su'o}? Or are you saying that using {lo} to mean {su'o} was a mistake?

I will repeat: before the constants fiasco, you *had* a proposal that

encompassed past usage, by making the meaning of {lo} unspecified.

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 05:13:36PM -0400, Rob Speer wrote:

> I want {lo} to literally mean {su'o} some of the time, because that's

> how you encompass past usage.

>

> So which does it mean?

By "literally mean {su'o}", you mean up to and including logical

quantification?

xorlo is definately not ever logically quantified.

I have yet to be convinced that the logical quantification of lo has

been relied upon in practice more than a handful of times, but then I

haven't finished studying Jordan's post.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 05:16:12PM -0400, Rob Speer wrote:

> I will repeat: before the constants fiasco, you *had* a proposal that

> encompassed past usage, by making the meaning of {lo} unspecified.

Explain to me the difference between the two, please.

If the meaning of lo is unspecified, then we clearly cannot allow

logical quantification or naku movement or whatever across, because we

have no idea what it means. So we're back at constants, as far as I can

tell.

-Robin



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 02:19:55PM -0700, Robin Lee Powell wrote:

> On Wed, Jun 09, 2004 at 05:16:12PM -0400, Rob Speer wrote:

> > I will repeat: before the constants fiasco, you *had* a proposal that

> > encompassed past usage, by making the meaning of {lo} unspecified.

>

> Explain to me the difference between the two, please.

>

> If the meaning of lo is unspecified, then we clearly cannot allow

> logical quantification or naku movement or whatever across, because we

> have no idea what it means. So we're back at constants, as far as I can

> tell.

What do you mean by logical quantification here? And what's so important about

naku movement that it's worth breaking the proposal for (given that it's

been used repeatedly as a justification for constants)?

--

Rob Speer



Posted by Anonymous on Wed 09 of June, 2004 21:47 GMT

On Wed, Jun 09, 2004 at 02:18:31PM -0700, Robin Lee Powell wrote:

> On Wed, Jun 09, 2004 at 05:13:36PM -0400, Rob Speer wrote:

> > I want {lo} to literally mean {su'o} some of the time, because that's

> > how you encompass past usage.

> >

> > So which does it mean?

>

> By "literally mean {su'o}", you mean up to and including logical

> quantification?

Sure. If the quantifiers aren't there, they should be unspecified, and it

should be possible for them to be {su'o lo ro} (by the old lo), which I think

is {su'o lo pa} by the new lo.

--

Rob Speer


I disagree


Posted by PierreAbbat on Wed 09 of June, 2004 22:05 GMT posts: 324

I am changing my vote to "no". I understand why we need an article for things that really broda but may not exist, such as unicorns, CCJ's fingernail, and the doctor I need. But if when applied to things that do exist, it violates the law of inverting the quantifier when {naku} passes through it, and something can be both inside and outside the box, I don't accept that.

phma



Posted by Anonymous on Thu 10 of June, 2004 02:53 GMT

On Wed, Jun 09, 2004 at 05:23:42PM -0400, Rob Speer wrote:

> On Wed, Jun 09, 2004 at 02:18:31PM -0700, Robin Lee Powell wrote:

> > On Wed, Jun 09, 2004 at 05:13:36PM -0400, Rob Speer wrote:

> > > I want {lo} to literally mean {su'o} some of the time, because

> > > that's how you encompass past usage.

> > >

> > > So which does it mean?

> >

> > By "literally mean {su'o}", you mean up to and including logical

> > quantification?

>

> Sure. If the quantifiers aren't there, they should be unspecified, and

> it should be possible for them to be {su'o lo ro} (by the old lo),

> which I think is {su'o lo pa} by the new lo.

So you would rather that "lo mlatu na fetsi" have the options of meaning

"No cats are female", correct?

I honestly don't care one way or the other. Optional, unspecified outer

quantifiers bothers me very little.

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:53 GMT

On Wed, Jun 09, 2004 at 03:05:23PM -0700, [email protected] wrote:

> I disagree

>

> I am changing my vote to "no". I understand why we need an article for

> things that really broda but may not exist, such as unicorns, CCJ's

> fingernail, and the doctor I need. But if when applied to things that

> do exist, it violates the law of inverting the quantifier when {naku}

> passes through it, and something can be both inside and outside the

> box, I don't accept that.

Nothing can be both inside and outside the box. Please read my other

posts. The context, and in particular the binding of "lo broda" can

vary between two statements, but that's not the same thing.

Aside from all that, as jatna I expect a counter-proposal along with

your No vote.

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 05:27:24PM -0400, Rob Speer wrote:

> On Wed, Jun 09, 2004 at 02:19:55PM -0700, Robin Lee Powell wrote:

> > On Wed, Jun 09, 2004 at 05:16:12PM -0400, Rob Speer wrote:

> > > I will repeat: before the constants fiasco, you *had* a proposal

> > > that encompassed past usage, by making the meaning of {lo}

> > > unspecified.

> >

> > Explain to me the difference between the two, please.

> >

> > If the meaning of lo is unspecified, then we clearly cannot allow

> > logical quantification or naku movement or whatever across, because

> > we have no idea what it means. So we're back at constants, as far

> > as I can tell.

>

> What do you mean by logical quantification here?

Quantification that can be affected by naku movemunt.

> And what's so important about naku movement that it's worth breaking

> the proposal for (given that it's been used repeatedly as a

> justification for constants)?

OK, I'm the wrong person to ask, but in CLL lo, the following are all

equivalent:

lo mlatu cu na fetsi

su'o da poi mlatu cu na fetsi

naku su'o da poi mlatu cu fetsi

Which means "It is not the case that there is at least on female cat".

In other words, if you actually use the quantifiers correctly, "lo mlatu

cu na fetsi" means "No cats are female".

It is absolutely trivial to find people using lo in a way that violates

this, and nearly impossible to find people using lo in a way that does

not violate this, where it applies, but Jordan doesn't seem to care.

lo skami ca'a na spofu — found on IRC in about twenty seconds of

searching. It is unlikely that the user meant to state that there

actually are *no* broken computers.

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Tue, Jun 08, 2004 at 08:57:39PM -0700, Jorge Llamb?as wrote:

> noras:

> > "ma'i la midju terdi la sadam na sai me la sauron" "In Middle

> > Earth-terms, Saddam is by no means a 'Sauron'."

> >

> > You really have 2 reference-frames wanted here: the real world and

> > Middle Earth. Saddam is in one, and the Sauron is in the other. I

> > read the lojban as setting the "world" to Middle Earth for the whole

> > sentence, in which case "Saddam" is meaningless. The Middle Earth

> > reference-frame applies only to 'Sauron'. So I suggest "la sadam na

> > sai me la sauron pe ma'i la midju terdi" = "Saddam is by no means a

> > 'Sauron'-of-middle-earth"

>

> The predicate here is {me la sauron}. Don't BAI tags add a place to

> the predicate,

Yes, but the only BAI is ma'i.

> so that we have "x1 is a 'Sauron' by standard x"?

Sort of.

me ME sumti to selbri

convert sumti to selbri/tanru element; x1 is

specific to sumti in aspect x2

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

Robin Lee Powell wrote:

>OK, I'm the wrong person to ask, but in CLL lo, the following are all

>equivalent:

>

>lo mlatu cu na fetsi

>

>su'o da poi mlatu cu na fetsi

>

>naku su'o da poi mlatu cu fetsi

>

>Which means "It is not the case that there is at least on female cat".

>

>In other words, if you actually use the quantifiers correctly, "lo mlatu

>cu na fetsi" means "No cats are female".

>

>

I thought the whole thing with na-movement was that quantifiers get

inverted when the pass a na-boundary. So {su'o da poi mlatu na fetsi}

becomes {naku roda fetsi}, i.e. "it is not the case that all cats are

female," which is what the naïve meaning of {lo mlatu na fetsi} is:

"some cat is not female".

Right?

~mark



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 08:07:47PM -0400, Mark E. Shoulson wrote:

> Robin Lee Powell wrote:

>

> >OK, I'm the wrong person to ask, but in CLL lo, the following are all

> >equivalent:

> >

> >lo mlatu cu na fetsi

> >

> >su'o da poi mlatu cu na fetsi

> >

> >naku su'o da poi mlatu cu fetsi

> >

> >Which means "It is not the case that there is at least on female

> >cat".

> >

> >In other words, if you actually use the quantifiers correctly, "lo

> >mlatu cu na fetsi" means "No cats are female".

>

> I thought the whole thing with na-movement was that quantifiers get

> inverted when the pass a na-boundary. So {su'o da poi mlatu na fetsi}

> becomes {naku roda fetsi}, i.e. "it is not the case that all cats are

> female," which is what the na??ve meaning of {lo mlatu na fetsi} is:

> "some cat is not female".

Nope. na broda == naku broda .

You're thinking of naku broda , which is another beast

entirely.

Or so John tells me.

-Rboin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

Mark E. Shoulson wrote:

>I thought the whole thing with na-movement was that quantifiers get

>inverted when the pass a na-boundary. So {su'o da poi mlatu na fetsi}

>becomes {naku roda fetsi}, i.e. "it is not the case that all cats are

>

>

Um. That's {naku ro mlatu cu fetsi}. Sorry.

>female," which is what the naïve meaning of {lo mlatu na fetsi} is:

>"some cat is not female".

>

>Right?

>

>~mark

>

>

>

>



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 08:17:15PM -0400, Mark E. Shoulson wrote:

> Mark E. Shoulson wrote:

>

> >I thought the whole thing with na-movement was that quantifiers get

> >inverted when the pass a na-boundary. So {su'o da poi mlatu na

> >fetsi} becomes {naku roda fetsi}, i.e. "it is not the case that all

> >cats are

> >

> >

> Um. That's {naku ro mlatu cu fetsi}. Sorry.

Except it's not.

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 03:38:07PM -0700, Robin Lee Powell wrote:

...

> It is absolutely trivial to find people using lo in a way that violates

> this, and nearly impossible to find people using lo in a way that does

> not violate this, where it applies, but Jordan doesn't seem to care.

Nope, I don't care.

Wrong usage is wrong usage. This was on the gotchas page back when

I first got involved lojban, and probably is still there.

I've tried to use it correctly in my translations. And recall at

least one occasion where I tried to get xorxes to fix a broken

sentence in le cmalu nolraitru (but he didn't want to .u'i).

> lo skami ca'a na spofu — found on IRC in about twenty seconds of

> searching. It is unlikely that the user meant to state that there

> actually are *no* broken computers.

IRC has a *lot* of incorrect usage. It's off the top of people's

heads, remember.

Believe it or not, I totally agree with xod that "past usage" cannot

be globally held up as a test, since the vast majority of past usage

is either incorrect or maybe-correct, pending bpfk descisions.

We shouldn't make frivolous changes---meaning changes to the language

as defined (as distinct from the language as used)---but priority

number one should be having things make sense (in particular in the

gadri system), and in some cases that'll potentially conflict with

past usage.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAx7C6DrrilS51AZ8RAo7IAKDMXKmx7wYTzXApy6zOZ7/o8BqEXACgugnK

hUggcm066otUTLSLAwr6mew=

=0Wz7

---END PGP SIGNATURE---



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 07:52:10PM -0500, Jordan DeLong wrote:

I had nothing useful to say to the rest; good points all

> Believe it or not, I totally agree with xod that "past usage" cannot

> be globally held up as a test, since the vast majority of past usage

> is either incorrect or maybe-correct, pending bpfk descisions.

>

> We shouldn't make frivolous changes---meaning changes to the language

> as defined (as distinct from the language as used)---but priority

> number one should be having things make sense (in particular in the

> gadri system), and in some cases that'll potentially conflict with

> past usage.

I actually disagreed with that prioritization. Past usage is, for me,

more important than making sense, for some value of "making sense".

-Robin



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 09:05:47AM -0700, Jorge Llambías wrote:

> Jordan:

> > > > Please note that this *drastically* changes the meaning of some

> > > > lojban sentences.

> > >

> > > Not any of the examples you have presented.

> > Currently {lo curnu cu na nenri le plise} means that *all* worms

> > in the entire world (if there are any worms) are not in the apple.

> > This meaning is not preserved if the naku can't move and make {su'o

> > lo curnu} into an assertion about {ro curnu}.

>

> Sorry, I thought you meant usage sentences. Usage sentences usually

> have a context so that such absolute terms hardly apply.

I'm not sure what you mean by this....

> > Do you agree that under xorlo {lo curnu cu nenri le plise} does not

> > imply that {naku lo curnu cu nenri le plise} is false?

>

> Give me context!

>

> {mi nenri le mi zdani} does not necessarily imply that

> {naku mi nenri le mi zdani} is false, I need to know if you want

> to keep every other possible circumstance constant or not.

This replies to Robin's post also:

Keep the context the same. Assume no CAhA trickery. Etc.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAx7dwDrrilS51AZ8RAlSgAJ9ZhocuGQMRhD/gqSLD5LaL+dEPVgCgq/EW

fyX4Obn7k+RFVLzniAVUlg4=

=foWg

---END PGP SIGNATURE---



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wed, Jun 09, 2004 at 10:46:00AM -0700, Jorge Llambías wrote:

> Jordan:

> > OTOH, I think all terms (excepting non-logical ones like ma) should

> > be definable in terms of quantified variables.

>

> Consider the term {zo lo}.

>

> zo lo cmavo

> "lo" is a cmavo

>

> mi ciska reno zo lo le papri

> I write 20 "lo"s on the page.

>

> Would it even cross you mind to think of instances of zo lo

> in the first sentence? The second sentence clearly deals with

> instances, the first one doesn't.

Quoted things can be a primative. They need to be considered in

the manner which names would have to be (some X such that X is that

particular thing).

I don't know whether quantification of "zo " works in the

manner you used it in the second sentence, and would suggest probably

not. It is grammatical, but probably because no one thought to go

out of their way to disalow it.

Check out x1 of lerfu. {la'e zo .abu} refers to a lerfu, but {naku

zo .abu lerfu}; zo .abu refers to the one and only lojban string

".abu".

To refer to the actualy letters on the page you don't want to use

quoted things---there's no quoted things on the page, only real

things. So {mi renoroi ciska zo lo le papri} seems better.

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAx7msDrrilS51AZ8RAvAyAJ4tylK8Ie3S02ZkURrYp2skUyVE6wCgrNQu

UOe8k3vGcYV915V2V/Glm70=

=pgY/

---END PGP SIGNATURE---



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wednesday 09 June 2004 18:16, Robin Lee Powell wrote:

> Nothing can be both inside and outside the box. Please read my other

> posts. The context, and in particular the binding of "lo broda" can

> vary between two statements, but that's not the same thing.

>

> Aside from all that, as jatna I expect a counter-proposal along with

> your No vote.

There has been way too much volume of discussion for me to digest it all, and

I still don't grok intensions, and I have some RL work to do, but here's a

try:

{lo broda} means one or more of those things that actually broda. If there are

none, which may be inherent in {broda} or may be indicated by another selbri

in the sentence, such as {claxu}, then it means one or more of those things

that actually broda, in some world where they exist.

Ex.: {la katr,in. claxu lo degji jgalu}. {lo degji jgalu} refers to the

fingernail which she has in some other world, which she lacks in this one.

{lo pavyseljirna cu blabi}. This means that in some world where unicorns

exist, there is at least one white one.

If one occurrence of {lo broda} is a sumti of more than one selbri, it means

the same one or more broda to all those selbri. If more than one {lo broda}

occurs, they can refer to different broda.

I'd like it to be intensional as well as quantified, but I don't know how to

word that. Can you write it better?

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT

On Wednesday 09 June 2004 21:30, Jordan DeLong wrote:

> Quoted things can be a primative. They need to be considered in

> the manner which names would have to be (some X such that X is that

> particular thing).

>

> I don't know whether quantification of "zo " works in the

> manner you used it in the second sentence, and would suggest probably

> not. It is grammatical, but probably because no one thought to go

> out of their way to disalow it.

>

> Check out x1 of lerfu. {la'e zo .abu} refers to a lerfu, but {naku

> zo .abu lerfu}; zo .abu refers to the one and only lojban string

> ".abu".

{zo .abu} is ungrammatical, IIRR. Both {zo} and {bu} try to grab {a} and

whichever one gets it turns it into the wrong kind of token for the other. zo

by. valsi .i me'o by. lerfu.

> To refer to the actualy letters on the page you don't want to use

> quoted things---there's no quoted things on the page, only real

> things. So {mi renoroi ciska zo lo le papri} seems better.

I see nothing wrong with {mi ciska reno zo lo le papri}.

phma

--

li fi'u vu'u fi'u fi'u du li pa



Posted by Anonymous on Thu 10 of June, 2004 02:54 GMT


> On Wed, Jun 09, 2004 at 08:37:56AM -0700, Jorge Llamb?as wrote:

> > By that you mean that {naku su'o lo curnu cu nenri}

> > cannot equal {su'o lo curnu naku nenri}, something that

> > I'm not disputing.

>

> But you do say that {naku lo curnu cu nenri} equals {lo curnu naku nenri},

Correct.

> and you say that {lo} can mean {su'o lo}, so there's a problem somewhere.

I don't say that. Bare {lo} is not quantified, so it can't possibly

mean {su'o lo}.

What can happen, and often does, is that a claim with {lo} is true and

so is a claim with {su'o lo}. But so what? Would you say that {la djan}

means {su'o prenu}? Because {naku la djan cu nenri} equals {la djan

naku nenri} too, and that says nothing about what happens if you

replace the unquantified term {la djan} with the quantified term

{su'o prenu}.

> What advantage does negation transparency give us that it's worth all this

> metaphysical debate to get?

I don't really understand what the metaphysical debate is about. The

proposal on {lo} has not changed since I presented it.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/


A few more items


Posted by noras on Thu 10 of June, 2004 02:59 GMT posts: 23

1. Three things on "la" example:

"la ci bakni po'u lo gusta pe la kaiapois cu mutce zabna ge lo vanbi gi lo cidja" = "The Three Cows Restaurant in Kaiapoi is a wonderful place, both atmosphere and food-wise."

a) 1st place of zabna is a favorable connotation of x2. So, the restaurant is a favorable connotation of environments and food? I think zabna is more correctly used with expressions (la'e zoi gy steadfast gy cu zanba la'e zo xarnu); that may be why it's so rarely used as the main brivla.

b) "ci bakni po'u lo gusta pe la kaiapois" is the name (Three Cows Which Are Restaurants of Kaiapoi). If you want just the "Three Cows", you need "la ci bakni ku po'u ..."

c) The use of "lo" in "lo gusta" disagrees with what's spelled out in the table. Per the table at the end, "la" means the speaker has a specific in mind, and "lo" means the the speaker does not have a specific in mind (in-mind = "No"). "po'u" means, effectively "poi du"; they are the same one(s). The speaker certainly know which restaurant it is if he knows it is the one in-mind from "la ci bakni". Either the example or the table needs to change; I suggest the table.

2. I'm not entirely comfortable with the "lo'i" example:

"ma cnano lo ka makau junta ce'u kei lo'i cifnu poi cazi jbena" = "What is the normal weight of a baby at childbirth?"

I think "cnano" is a mathematical thing - average/mean; that means "ni" would be better than "ka". Also, I read past-tense, pronounced like "red" "lo'i cifnu poi cazi jbena" as "a/some set(s) of babies who were just now born"; this is certainly not the only possible meaning, but would prefer (not insist on) something clearer. Perhaps "pe ca lo nu jbena"?

3. In "le'i" example:

"ro le verba pu cuxna pa karda le'i cnita selcra" = "Each child chose a card from the face-down collection."

x1 crane x2 means x1 is in front (or is THE front) of x2. So selcra is something that has a front. "cnita selcra", to me, means "beneath type of thing that has a front; this doesn't bring to mind "face-down". How about "le'i seke cnita crane". This would be the x2 of "cnita crane" (x1 cnita crane x2 = "x1 is a beneath-type-of-front of x2"); if would be the something that has a beneath-type-of-front.

mi'e noras.



Posted by Anonymous on Thu 10 of June, 2004 04:09 GMT

At 01:38 PM 6/9/04 -0700, Robin wrote:

>For any *fixed* context, including the binding of the constant (which is

>only in the speaker's head), "lo curnu cu nenri le plise" implies that

>"naku lo curnu cu nenri le plise" is false (assuming I understand naku,

>which is false).

>

>However, there is nothing the requires the speaker to maintain the

>binding of the constant from one statement to the next, and that's where

>the problem is.

>

>There is nothing wrong with the following interchange:

>

>A: lo curnu cu nenri le plise

>

>B: .i ma curnu

>

>A: ro curnu poi nenri le plise .iji'a naku lo curnu cu nenri le plise

>

>B: .i .oisai ma curnu

>

>A: ro curnu poi na nenri le plise

>

>My question is, is "lo curnu cu nenri le plise" false if there is, in

>fact, no worm in the apple? I believe that it is false in that case,

>but I'd like others' opinions. Assuming I'm right, the above exchange

>is only sensible if there is at least one worm in the apple.

>

>Now, if it was "lo curnu cu nenri *lo* plise", that would be different.

Picked this message at random. This is not a comment on xorlo.

Your example "ma curnu" suggests to me the negation example of "The present

king of France is bald" vs The present king of France is not bald".

At which point I'll ask (to fit the example) if your questions make sense

(and have answers) if you replace curnu by crida (I'm thinking maybe Tinker

Bell here, but I'm likewise thinking "unicorn" and "present king of France".

lojbab

--

lojbab [email protected]

Bob LeChevalier, Founder, The Logical Language Group

(Opinions are my own; I do not speak for the organization.)

Artificial language Loglan/Lojban: http://www.lojban.org



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

On Wed, Jun 09, 2004 at 03:14:50PM -0700, Robin Lee Powell wrote:

> So you would rather that "lo mlatu na fetsi" have the options of meaning

> "No cats are female", correct?

Certainly.

Though doesn't it mean that in Constant Land, also, because it says that

Mr. Cat is not female u'i? Not that I'm warming up to the idea of Mr. Cat,

I'm just saying that I don't think this is too far from what you and xorxes

want anyway.

--

Rob Speer



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

On Thu, Jun 10, 2004 at 12:28:53AM -0400, Rob Speer wrote:

> On Wed, Jun 09, 2004 at 03:14:50PM -0700, Robin Lee Powell wrote:

> > So you would rather that "lo mlatu na fetsi" have the options of

> > meaning "No cats are female", correct?

>

> Certainly.

>

> Though doesn't it mean that in Constant Land, also, because it says

> that Mr. Cat is not female u'i? Not that I'm warming up to the idea

> of Mr. Cat, I'm just saying that I don't think this is too far from

> what you and xorxes want anyway.

It *can* mean that, yes, but by a completely different mechanism.

-Robin



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

On Wed, Jun 09, 2004 at 07:46:16PM -0700, Jorge Llamb?as wrote:

> I don't say that. Bare {lo} is not quantified, so it can't possibly

> mean {su'o lo}.

I was sold on the idea that bare {lo} would be unspecified in such a way that

it would cover several different usages of lo, including {su'o} and

intensionality. This was the very thing that convinced me that the proposal

would work.

Since this was apparently not true, the proposal now has my NO vote.

You and Robin even seem to be talking about slightly different versions of the

proposal, and I'd vote for Robin's version. I think you're asking for too much

with your proposal, and you need to back off to something that more people can

understand and accept.

--

Rob Speer



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

Rob Speer scripsit:

> On Wed, Jun 09, 2004 at 07:46:16PM -0700, Jorge Llamb?as wrote:

> > I don't say that. Bare {lo} is not quantified, so it can't possibly

> > mean {su'o lo}.

>

> I was sold on the idea that bare {lo} would be unspecified in such

> a way that it would cover several different usages of lo, including

> {su'o} and intensionality. This was the very thing that convinced me

> that the proposal would work.

>

> Since this was apparently not true, the proposal now has my NO vote.

I hate to say this, but the problem is with the meaning of "meaning".

When xorxes sas above that "lo" doesn't mean "su'o lo", he means that

they are not *synonymous*, precisely because "lo" covers different cases

that "su'o lo" does not. However, as xorxes says in another posting,

"lo" and "su'o lo" can both apply.

--

All Gaul is divided into three parts: the part John Cowan

that cooks with lard and goose fat, the part www.ccil.org/~cowan

that cooks with olive oil, and the part that www.reutershealth.com

cooks with butter. — David Chessler [email protected]



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

John Cowan wrote:

>Rob Speer scripsit:

>

>

>

>>On Wed, Jun 09, 2004 at 07:46:16PM -0700, Jorge Llamb?as wrote:

>>

>>

>>>I don't say that. Bare {lo} is not quantified, so it can't possibly

>>>mean {su'o lo}.

>>>

>>>

>>I was sold on the idea that bare {lo} would be unspecified in such

>>a way that it would cover several different usages of lo, including

>>{su'o} and intensionality. This was the very thing that convinced me

>>that the proposal would work.

>>

>>Since this was apparently not true, the proposal now has my NO vote.

>>

>>

>

>I hate to say this, but the problem is with the meaning of "meaning".

>

>When xorxes sas above that "lo" doesn't mean "su'o lo", he means that

>they are not *synonymous*, precisely because "lo" covers different cases

>that "su'o lo" does not. However, as xorxes says in another posting,

>"lo" and "su'o lo" can both apply.

Are there any examples (using XS lo) where lo broda nau brode .iku'i

su'oda broda nau na brode?

--

Motorists honked in celebration in this Ramadi as news spread of the

assassination of the president of the Iraqi Governing Council Ezzidin Salim

Monday. "The GC is nothing," one man shouted. "They are not the Governing

Council. They are the Prostitution Council."



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

xod scripsit:

> Are there any examples (using XS lo) where lo broda nau brode .iku'i

> su'oda broda nau na brode?

Sure: lo -unicorn -has-a-horn. su'o da poi -unicorn -has-a-horn

would be false, because it says "there exists at least one thing

which is a unicorn and it has a horn", but there are no unicorns.

--

Knowledge studies others / Wisdom is self-known; John Cowan

Muscle masters brothers / Self-mastery is bone; [email protected]

Content need never borrow / Ambition wanders blind; www.ccil.org/~cowan

Vitality cleaves to the marrow / Leaving death behind. --Tao 33 (Bynner)



Posted by Anonymous on Thu 10 of June, 2004 06:08 GMT

John Cowan wrote:

>xod scripsit:

>

>

>

>>Are there any examples (using XS lo) where lo broda nau brode .iku'i

>>su'oda broda nau na brode?

>>

>>

>

>Sure: lo -unicorn -has-a-horn. su'o da poi -unicorn -has-a-horn

>would be false, because it says "there exists at least one thing

>which is a unicorn and it has a horn", but there are no unicorns.

>

>

The entire context must be one in which unicorns exist; otherwise the

sentences are not even false but meaningless.

--

Motorists honked in celebration in this Ramadi as news spread of the assassination of the president of the Iraqi Governing Council Ezzidin Salim Monday. "The GC is nothing," one man shouted. "They are not the Governing Council. They are the Prostitution Council."



Posted by Anonymous on Thu 10 of June, 2004 17:47 GMT

xod:

> Are there any examples (using XS lo) where lo broda nau brode .iku'i

> su'oda broda nau na brode?

lo mikce nau se nitcu mi iku'i naku su'o mikce nau se nitcu mi

mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 10 of June, 2004 17:47 GMT

John Cowan:

> When xorxes sas above that "lo" doesn't mean "su'o lo", he means that

> they are not *synonymous*, precisely because "lo" covers different cases

> that "su'o lo" does not. However, as xorxes says in another posting,

> "lo" and "su'o lo" can both apply.

Right.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 10 of June, 2004 17:47 GMT

noras:

> A few more items

> 1. Three things on "la" example:

> "la ci bakni po'u lo gusta pe la kaiapois cu mutce zabna ge lo vanbi gi lo

> cidja" = "The Three Cows Restaurant in Kaiapoi is a wonderful place, both

> atmosphere and food-wise."

>

> a) 1st place of zabna is a favorable connotation of x2. So, the restaurant

> is a favorable connotation of environments and food? I think zabna is more

> correctly used with expressions (la'e zoi gy steadfast gy cu zanba la'e zo

> xarnu); that may be why it's so rarely used as the main brivla.

Right, I forgot about that. I was using my definition of zabna.

See mabla

So, how do we say:

"x1 is excellent/wonderful/great/first-class/top-quality

in property x2 by standard x3"

in a more acceptable canonical form?

> b) "ci bakni po'u lo gusta pe la kaiapois" is the name (Three Cows Which Are

> Restaurants of Kaiapoi). If you want just the "Three Cows", you need "la ci

> bakni ku po'u ..."

Good point.

> c) The use of "lo" in "lo gusta" disagrees with what's spelled out in the

> table. Per the table at the end, "la" means the speaker has a specific in

> mind, and "lo" means the the speaker does not have a specific in mind

> (in-mind = "No"). "po'u" means, effectively "poi du"; they are the same

> one(s). The speaker certainly know which restaurant it is if he knows it is

> the one in-mind from "la ci bakni". Either the example or the table needs to

> change; I suggest the table.

I understood that as saying that in-mindedness was not an issue for

the lo-series, not that you cannot have something in mind when you

use lo. Any suggestions for a better wording?

> 2. I'm not entirely comfortable with the "lo'i" example:

> "ma cnano lo ka makau junta ce'u kei lo'i cifnu poi cazi jbena" = "What is

> the normal weight of a baby at childbirth?"

>

> I think "cnano" is a mathematical thing - average/mean; that means "ni" would

> be better than "ka".

The x1 of junta, as signaled by makau, is what's averaged. If I used

{ni}, we would have a second order amount coming into play: the extent to

which some weight is the weight of a baby.

> Also, I read past-tense, pronounced like "red" "lo'i

> cifnu poi cazi jbena" as "a/some set(s) of babies who were just now born";

> this is certainly not the only possible meaning, but would prefer (not insist

> on) something clearer. Perhaps "pe ca lo nu jbena"?

We do tend to read {ca} as {nau}, that's true. I would like

to leave the example as counterweight for that bias, if there

is not a lot of opposition.

> 3. In "le'i" example:

> "ro le verba pu cuxna pa karda le'i cnita selcra" = "Each child chose a card

> from the face-down collection."

>

> x1 crane x2 means x1 is in front (or is THE front) of x2. So selcra is

> something that has a front. "cnita selcra", to me, means "beneath type of

> thing that has a front; this doesn't bring to mind "face-down". How about

> "le'i seke cnita crane". This would be the x2 of "cnita crane" (x1 cnita

> crane x2 = "x1 is a beneath-type-of-front of x2"); if would be the something

> that has a beneath-type-of-front.

My thinking was "beneath faced", {selcra be lo cnita}, something whose

face is beneath, not {cnita je selcra}.

ki'e mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 10 of June, 2004 17:47 GMT

The token-type distinction is a very different critter from whatever it is that you are working with. I am not sure it is even analogous. To be sure, Lojban doesn't deal with it at all (not merely not well) and to that extent, it may be like your problem (what is that by the way?).

wrote:

Jordan:

> OTOH, I think all terms (excepting non-logical ones like ma) should

> be definable in terms of quantified variables.

Consider the term {zo lo}.

zo lo cmavo

"lo" is a cmavo

mi ciska reno zo lo le papri

I write 20 "lo"s on the page.

Would it even cross you mind to think of instances of zo lo

in the first sentence? The second sentence clearly deals with

instances, the first one doesn't.

mu'o mi'e xorxes

'__

Do you Yahoo!?

Friends. Fun. Try the all-new Yahoo! Messenger.

http://messenger.yahoo.com/



Posted by Anonymous on Thu 10 of June, 2004 17:47 GMT

I am unclear about which feature you consider necessary and which a wart. They are different things for me. That external quantifiers quantify over individuals seems necessary, but (therefore, in fact) not a wart. That internal quantifiers set up -ads rather than just so-and-so mmany individuals seems a wart and not at all necessary (as witness the several other ways that it has been done or have been proposed).

John E Clifford scripsit:

> > > {ci gerku cu batci le pa bongu}.

> >

> > Which is to say, "ci [lo pa] gerku".

> >

> Lord, I hope not: singletons don't bite bones (I am not sure about

> triads, but probably not). I suspect this is just a terminological

> problems, though working out the logic is a bitch — because of the

> quantifiers, as usual.

When there is an outer quantifier to lo, but no inner quantifier, it

retains its meaning of quantifying over individuals, for the sake of

backward compatibility. This is a wart, but a necessary one.

--

We call nothing profound [email protected]

that is not wittily expressed. John Cowan

--Northrop Frye (improved) http://www.reutershealth.com



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

On Wed, Jun 09, 2004 at 09:38:34PM -0400, Pierre Abbat wrote:

> On Wednesday 09 June 2004 21:30, Jordan DeLong wrote:

> > Quoted things can be a primative. They need to be considered in

> > the manner which names would have to be (some X such that X is that

> > particular thing).

> >

> > I don't know whether quantification of "zo " works in the

> > manner you used it in the second sentence, and would suggest probably

> > not. It is grammatical, but probably because no one thought to go

> > out of their way to disalow it.

> >

> > Check out x1 of lerfu. {la'e zo .abu} refers to a lerfu, but {naku

> > zo .abu lerfu}; zo .abu refers to the one and only lojban string

> > ".abu".

>

> {zo .abu} is ungrammatical, IIRR. Both {zo} and {bu} try to grab {a} and

> whichever one gets it turns it into the wrong kind of token for the other. zo

> by. valsi .i me'o by. lerfu.

{me'o by.} is not a lerfu, it's a mekso. {la'e zo by. lerfu}.

You're right that {zo .abu} is wrong, but it's not relevant to the

point I was making.... (replace all the .abu with by. and read it

again).

> > To refer to the actualy letters on the page you don't want to use

> > quoted things---there's no quoted things on the page, only real

> > things. So {mi renoroi ciska zo lo le papri} seems better.

>

> I see nothing wrong with {mi ciska reno zo lo le papri}.

Care to give a reason why you don't think my above reasoning makes

sense?

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAyHr1DrrilS51AZ8RAoG7AJ4pLsAYXcK6NxjEcd7iy5R+77GCJACgqTRK

ObatuZVIuCPyVZYd2ewylhk=

=NP/q

---END PGP SIGNATURE---



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

I think that masses (at least in the {loi} sense) have a different set of rules nowadays and a more useful one. This is getting on toward the species sense (without using {lo}!). and there, at at least one level, the claim {broda gi'enai broda} holds. But, like all claims at this level, it is merely an idiom, a convenient way to say a rather complicated thing, in this case that the node in question spans the break between broda and naku broda.

[email protected] wrote:Robin Lee Powell scripsit:

> What about the mass gadri? I thought it was part of the lore of masses

> that they had all the features of their members, such that the mass of

> all lions is both male and female?

Very true. So "loi cinfo cu nakni" is true (and "loi cinfo cu na nakni"

is false) and "loi cinfo cu fetsi" is true (and "loi cinfo cu na fetsi"

is false). This is not a *contradiction*, just surprising.

--

John Cowan [email protected] www.ccil.org/~cowan

Female celebrity stalker, on a hot morning in Cairo:

"Imagine, Colonel Lawrence, ninety-two already!"

El Auruns's reply: "Many happy returns of the day!"



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

I've decided to vote "no".

My reasoning is that xorlo isn't well-defined, but oldlo very much

is (in fact, it's the best defined article out of all of them).

--

Jordan DeLong

[email protected]

-- Attached file included as plaintext by Ecartis --

---BEGIN PGP SIGNATURE---

Version: GnuPG v1.2.1 (FreeBSD)

iD8DBQFAyHt8DrrilS51AZ8RAudGAJ45Hu/Y77lDV8qUw/h4BYXx0VUMOgCeOosZ

I0Ory8By5vdRelXlZgYLQ8I=

=o8vQ

---END PGP SIGNATURE---



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

If there are quantifiers in the offing at all, whether unsated or inspecific or whateve, then not e that all of them except {no} and {me'i} entail {su'o} . (I hope old {su'o lo} is not proposed {su'o lo pa}; I want to quantify over things, not sets or groups or,,, of them.

Rob Speer wrote:On Wed, Jun 09, 2004 at 02:18:31PM -0700, Robin Lee Powell wrote:

> On Wed, Jun 09, 2004 at 05:13:36PM -0400, Rob Speer wrote:

> > I want {lo} to literally mean {su'o} some of the time, because that's

> > how you encompass past usage.

> >

> > So which does it mean?

>

> By "literally mean {su'o}", you mean up to and including logical

> quantification?

Sure. If the quantifiers aren't there, they should be unspecified, and it

should be possible for them to be {su'o lo ro} (by the old lo), which I think

is {su'o lo pa} by the new lo.

--

Rob Speer



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

Wrong. In spite of where it appears in the sentence (next to the predicate), {na} negates the whole sentence, i.e., is equivalent to {naku} at the very front. On the other hand, repalacing the {na} with {naku} right before the predicate means that negation has moved from the front to this internal location and so all intervening quantifiers change. I gather that it is this fact — and the comnfusion it engenders among the careless — that has led to the proposal to make {lo} — unlike anything else in the language — negation-transparent. It doesn't work.

"Mark E. Shoulson" wrote:Robin Lee Powell wrote:

>OK, I'm the wrong person to ask, but in CLL lo, the following are all

>equivalent:

>

>lo mlatu cu na fetsi

>

>su'o da poi mlatu cu na fetsi

>

>naku su'o da poi mlatu cu fetsi

>

>Which means "It is not the case that there is at least on female cat".

>

>In other words, if you actually use the quantifiers correctly, "lo mlatu

>cu na fetsi" means "No cats are female".

>

>

I thought the whole thing with na-movement was that quantifiers get

inverted when the pass a na-boundary. So {su'o da poi mlatu na fetsi}

becomes {naku roda fetsi}, i.e. "it is not the case that all cats are

female," which is what the naïve meaning of {lo mlatu na fetsi} is:

"some cat is not female".

Right?

~mark



Posted by Anonymous on Thu 10 of June, 2004 17:48 GMT

Jordan:

> > > Check out x1 of lerfu. {la'e zo .abu} refers to a lerfu, but {naku

> > > zo .abu lerfu}; zo .abu refers to the one and only lojban string

> > > ".abu".

(BTW, that definition of lerfu should be fixed at some point, it seems

to suggest that {zo a bu} is grammatical. Or maybe it will end up being

grammatical, we still have to figure out magic words.)

I wonder what you would understand of this dialog:

A: ry