[lojban] Re: interactions between tenses, other tenses, and NA

6 views
Skip to first unread message

jjllambias2000

unread,
Sep 27, 2002, 10:48:47 AM9/27/02
to loj...@yahoogroups.com
la djan cusku di'e

> The principle is that everything is exported
> to the prenex in the order in which it (first) appears, *except* NA,
> which is always exported to the very beginning. In that way,
inserting
> "na" before the selbri (mixed with tenses any way you like) is
always the
> exact contradictory negation of the version without "na".

I doubt this can work in practice. {na} is consistently misused
by almost everybody, so I suspect that the rule will be eventually
generalized to order of appearance, {na} not excepted.

> --
> John Cowan jcowan@r... www.reutershealth.com www.ccil.org/~cowan
> "The exception proves the rule." Dimbulbs think: "Your
counterexample proves
> my theory." Classicists think "'Probat' means 'tests': the
exception puts the
> rule to the proof." But legal historians know it means "Evidence
for an
> exception is evidence of the existence of a rule in cases not
excepted from."

In Spanish the saying is "la excepci�n que confirma la regla",
so the classisists' version is not possible for us. (The legal
historians' version is more difficult to explain to the dimbulbs.)

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->
Sell a Home for Top $
http://us.click.yahoo.com/RrPZMC/jTmEAA/MVfIAA/GSaulB/TM
---------------------------------------------------------------------~->

To unsubscribe, send mail to lojban-un...@onelist.com

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

And Rosta

unread,
Sep 27, 2002, 3:52:15 PM9/27/02
to loj...@yahoogroups.com
Jorge:

> la djan cusku di'e
>
> > The principle is that everything is exported
> > to the prenex in the order in which it (first) appears, *except* NA,
> > which is always exported to the very beginning. In that way,
> inserting
> > "na" before the selbri (mixed with tenses any way you like) is
> always the
> > exact contradictory negation of the version without "na".
>
> I doubt this can work in practice. {na} is consistently misused
> by almost everybody, so I suspect that the rule will be eventually
> generalized to order of appearance, {na} not excepted.

Perhaps we should make a conscious decision not to correct such
misuse, so as to encourage the 'correct' rule to die from neglect?

--And.

And Rosta

unread,
Sep 27, 2002, 7:48:26 PM9/27/02
to loj...@yahoogroups.com
jorge:
> If I understand correctly, "the usual language of formal
> logic" would have something like ~Fab
>
> This can be described as:
>
> 1- Negation in front of the predicate
> 2- Negation in front of the whole expression
>
> Lojban does: a~Fb, so as far as negation goes, it either follows
> the usual language of formal logic (by 1) or it does not follow the
> usual language of formal logic (by 2).

But try adding a quantifier. Ax~Fxb. But in Lojban {ro da na brode
ko'a} means ~AxFxb. So there is no way that Lojban follows the
language of formal logic.

--And.

------------------------ Yahoo! Groups Sponsor ---------------------~-->

Plan to Sell a Home?
http://us.click.yahoo.com/J2SnNA/y.lEAA/MVfIAA/GSaulB/TM

And Rosta

unread,
Sep 27, 2002, 3:52:19 PM9/27/02
to loj...@yahoogroups.com
John:
> > Of course, and everybody says things wrong every now and then
> > even in their native language. All I'm saying is that the rule
> > for {na} is extremely difficult to master, at least for me.
>
> It is pretty alien, yes.

To me it seems one of the more naturalistic features of Lojban --
a quirky, exceptional, counterintuitive, unnecessary complication,
of the sort natlangs are full of & Lojban is largely free of.

--And.

------------------------ Yahoo! Groups Sponsor ---------------------~-->

Sell a Home with Ease!
http://us.click.yahoo.com/SrPZMC/kTmEAA/MVfIAA/GSaulB/TM

Jim Carter

unread,
Sep 27, 2002, 12:19:38 PM9/27/02
to loj...@yahoogroups.com
On Fri, 27 Sep 2002, jjllambias2000 wrote:
> la djorden cusku di'e
> > It's actually pretty simple: there's no need to do real thinking
> > about the sentence:
> > - if you can rephrase it as "It is false that: foo", the na is
> fine.
>
> I call that real thinking, and that't exactly what I do to
> analyze it. But I don't want to have to rephrase a sentence
> in order to understand it. I don't want to have to translate
> it in the first place.

Perhaps the real issue is that you have to reprogram your semantic analyser
for real logic. Mapping Lojban 1-1 into an illogical natlang is going to
mangle the result, particularly where "carbon units" are most sloppy in
their logic. I've actually gotten into the habit of using Lojban-style
predicate logic and quantifier scope (and negation) internally, I remember
a case where two friends were discussing, and one said the exact opposite
of what he intended due to a botched negation, and I was able to catch this
on the fly and "translate", short-circuiting 15 minutes of hot air.

James F. Carter Voice 310 825 2897 FAX 310 206 6673
UCLA-Mathnet; 6115 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA 90095-1555
Email: ji...@math.ucla.edu http://www.math.ucla.edu/~jimc (q.v. for PGP key)


------------------------ Yahoo! Groups Sponsor ---------------------~-->

John Cowan

unread,
Sep 27, 2002, 4:11:36 PM9/27/02
to And Rosta, loj...@yahoogroups.com
And Rosta scripsit:

> To me it seems one of the more naturalistic features of Lojban --
> a quirky, exceptional, counterintuitive, unnecessary complication,
> of the sort natlangs are full of & Lojban is largely free of.

What is alien is that a contradictory negation particle should be other
than at the beginning of the sentence.

In Loglan, sentence-initial "no" served this function. IIRC, Lojbab
consciously moved it from the natural sentence-initial position to just
before the selbri "because it was more naturalistic". IMHO a mistake.

--
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 jco...@reutershealth.com

------------------------ Yahoo! Groups Sponsor ---------------------~-->

John Cowan

unread,
Sep 26, 2002, 11:21:18 PM9/26/02
to ara...@newmail.net, lojba...@lojban.org
Adam Raizen scripsit:

> 'na' can be interspersed among tenses, and I thought that this was to
> allow constructions such as 'roroi na' = 'always not', 'ka'e na' = 'is
> capable of not', etc. If NA must move to the front of the prenex, but
> tenses don't, then what is the point of having NA interspersed with
> tenses?

There is no *point*, but it's not forbidden either.

> Therefore, I think that tenses work the same as NA,

That was not the intent.

> Does anyone object to this? Could anyone *really* think that 'roroi na
> broda' means 'not always brodas' (i.e. 'sometimes doesn't broda')? Can
> we get a (quasi-)official pronouncement from Cowan?

Yes, that is what it means. The principle is that everything is exported


to the prenex in the order in which it (first) appears, *except* NA,
which is always exported to the very beginning. In that way, inserting
"na" before the selbri (mixed with tenses any way you like) is always the

exact contradictory negation of the version without "na". (Exception:
when the selbri is a GOhA that has a "na" semantically embedded in it, in
which case the added "na" is pleonastic.)

To say what you want to say with "roroi na", use "roroi naku" instead.

Quasi-officially yours,

--
John Cowan jco...@reutershealth.com www.reutershealth.com www.ccil.org/~cowan

And Rosta

unread,
Sep 27, 2002, 7:33:54 PM9/27/02
to loj...@yahoogroups.com
jimc:

> On Fri, 27 Sep 2002, jjllambias2000 wrote:
> > la djorden cusku di'e
> > > It's actually pretty simple: there's no need to do real thinking
> > > about the sentence:
> > > - if you can rephrase it as "It is false that: foo", the na is
> > fine.
> >
> > I call that real thinking, and that't exactly what I do to
> > analyze it. But I don't want to have to rephrase a sentence
> > in order to understand it. I don't want to have to translate
> > it in the first place.
>
> Perhaps the real issue is that you have to reprogram your semantic analyser
> for real logic. Mapping Lojban 1-1 into an illogical natlang is going to
> mangle the result, particularly where "carbon units" are most sloppy in
> their logic. I've actually gotten into the habit of using Lojban-style
> predicate logic and quantifier scope (and negation) internally,

Right, but the whole problem here is that the iinear/one-dimensionsal
hierarchy of scope relations is, in a highly intuitive way, reflected
in Lojban by linear order *except* for bare {na}. So when processing
Lojban logically you have to remember two facts:

Fact A. Scope goes left to right, as you would expect.
Fact B. {na} is an exception.

Fact B is counterintuitive, and it is an effort to remember to
remember it.

--And.

------------------------ Yahoo! Groups Sponsor ---------------------~-->

4 DVDs Free +s&p Join Now
http://us.click.yahoo.com/pt6YBB/NXiEAA/MVfIAA/GSaulB/TM

And Rosta

unread,
Sep 27, 2002, 8:45:06 PM9/27/02
to lojba...@lojban.org
Jordan DeLong
> On Sat, Sep 28, 2002 at 12:42:40AM +0100, And Rosta wrote:
> > John:

> > > And Rosta scripsit:
> > > > To me it seems one of the more naturalistic features of Lojban --
> > > > a quirky, exceptional, counterintuitive, unnecessary complication,
> > > > of the sort natlangs are full of & Lojban is largely free of.
> > >
> > > What is alien is that a contradictory negation particle should be other
> > > than at the beginning of the sentence.
> > >
> > > In Loglan, sentence-initial "no" served this function. IIRC, Lojbab
> > > consciously moved it from the natural sentence-initial position to just
> > > before the selbri "because it was more naturalistic". IMHO a mistake.
> >
> > As is obvious, I agree that the decision was a terrible mistake, but
> > the idea that "it was more naturalistic" is fairly defensible,
> > given that (a) some lects of English have it, and (b) quirkiness,
> > exception-riddenness, counterintuiveness and unnecessary complication
> > is highly characteristic of natlangs, as evidenced by the way that
> > more accomplished naturalistic artlangers deliberately try to add
> > it to their conlangs.
>
> There's nothing to prevent you from always saying "naku" or always
> putting it at the front of the bridi. (That is, unless the fact
> that you never seem to produce any lojban text could be considered
> preventative.)

My point, as always, is about good and bad design, not about
infringements of my linguistic rights or unwelcomely chafing
constraints on my usage.

See http://nuzban.wiw.org/wiki/index.php?And%20Rosta in order to get
a better understanding of the background to the technical debates on
Lojban list.

--And.


py...@aol.com

unread,
Sep 28, 2002, 4:31:55 AM9/28/02
to loj...@yahoogroups.com
In a message dated 9/27/2002 3:39:31 PM Central Daylight Time, jjlla...@hotmail.com writes:

<<
If I understand correctly, "the usual language of formal
logic" would have something like ~Fab

This can be described as:

1- Negation in front of the predicate
2- Negation in front of the whole expression

Lojban does: a~Fb, so as far as negation goes, it either follows
the usual language of formal logic (by 1) or it does not follow the
usual language of formal logic (by 2).

>>
Officially the Lojban is type 2 -- negation before the whole expression -- that being the usual meaning of the corresponding logical form.  However, in those logics that do allow the usual Lojban form, the meaning is usually 1.  But even then the internal negation form is said not to be usual, but wehipped out for special purposes (to disambiguate sentences involving Ruwwellian descriptions or for possibly empty worlds).
Yahoo! Groups Sponsor
ADVERTISEMENT

To unsubscribe, send mail to lojban-un...@onelist.com

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Lionel Vidal

unread,
Sep 28, 2002, 9:16:50 AM9/28/02
to lojba...@lojban.org
and:

> See http://nuzban.wiw.org/wiki/index.php?And%20Rosta in order to get
> a better understanding of the background to the technical debates on
> Lojban list.

That is something I did not bothered to do till recently, mainly because
I did not realise it exists, but after some Wiki exploring, I encourage
everybody to read the presentation pages on and, xorxes, Jay and all
the people whose name comes so often in that list. I found it very
usefull to help me understand the positions and points of view
expressed herein.

mu'omi'e lioNEL


jjllambias2000

unread,
Sep 27, 2002, 4:38:45 PM9/27/02
to loj...@yahoogroups.com

la pycyn cusku di'e

> As usual, "logical" applied to Lojban means "how to do it in the
usual
> language of formal logic," which, in this case (as usually), is the
standrad
> Lojban convention.

If I understand correctly, "the usual language of formal
logic" would have something like ~Fab

This can be described as:

1- Negation in front of the predicate
2- Negation in front of the whole expression

Lojban does: a~Fb, so as far as negation goes, it either follows
the usual language of formal logic (by 1) or it does not follow the
usual language of formal logic (by 2).

>(But there are perfectly good logical systems tht do it
> otherwise -- including especially ones that, like Lojban, are SVO
rather than
> VSO.)

Right.

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->

Home Selling? Try Us!
http://us.click.yahoo.com/QrPZMC/iTmEAA/MVfIAA/GSaulB/TM
---------------------------------------------------------------------~->

To unsubscribe, send mail to lojban-un...@onelist.com

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

jjllambias2000

unread,
Sep 27, 2002, 12:03:15 PM9/27/02
to loj...@yahoogroups.com
la djorden cusku di'e

> It's actually pretty simple: there's no need to do real thinking
> about the sentence:
> - if you can rephrase it as "It is false that: foo", the na is
fine.

I call that real thinking, and that't exactly what I do to
analyze it. But I don't want to have to rephrase a sentence
in order to understand it. I don't want to have to translate
it in the first place.

> - if not, you can make it a "naku" and get position defined scope
> - also consider a cmavo from NAhE, which is frequently better
anyway.

Yes, that's a possibility, not use {na}. But you still have to
cope with other people's use. First you have to figure out what
they actually said , and then you have to decide how likely
it is that they might have meant something else. (Although often
you can figure out what they meant independently of what they
actually said.)

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->

And Rosta

unread,
Sep 27, 2002, 7:42:40 PM9/27/02
to John Cowan, loj...@yahoogroups.com
John:

> And Rosta scripsit:
>
> > To me it seems one of the more naturalistic features of Lojban --
> > a quirky, exceptional, counterintuitive, unnecessary complication,
> > of the sort natlangs are full of & Lojban is largely free of.
>
> What is alien is that a contradictory negation particle should be other
> than at the beginning of the sentence.
>
> In Loglan, sentence-initial "no" served this function. IIRC, Lojbab
> consciously moved it from the natural sentence-initial position to just
> before the selbri "because it was more naturalistic". IMHO a mistake.

As is obvious, I agree that the decision was a terrible mistake, but


the idea that "it was more naturalistic" is fairly defensible,
given that (a) some lects of English have it, and (b) quirkiness,
exception-riddenness, counterintuiveness and unnecessary complication
is highly characteristic of natlangs, as evidenced by the way that
more accomplished naturalistic artlangers deliberately try to add
it to their conlangs.

--And.

------------------------ Yahoo! Groups Sponsor ---------------------~-->

And Rosta

unread,
Sep 27, 2002, 8:37:52 PM9/27/02
to no...@lojban.org, lojba...@lojban.org
Noras:
> la xorxes cusku di'e:

> >la djan cusku di'e
> >
> > > The principle is that everything is exported
> > > to the prenex in the order in which it (first) appears, *except* NA,
> > > which is always exported to the very beginning. In that way,
> >inserting
> > > "na" before the selbri (mixed with tenses any way you like) is
> >always the
> > > exact contradictory negation of the version without "na".
> >
> >I doubt this can work in practice. {na} is consistently misused
> >by almost everybody, so I suspect that the rule will be eventually
> >generalized to order of appearance, {na} not excepted.
>
> "na" exporting to first in the prenex is necessary for "na go'i" to
> work. "na go'i" MUST be the exact negation of the "xu" sentence to allow
> answering questions to work.
>
> xu do roroi klama le zarci
> ..i na go'i
> You don't want "na go'i" to mean you always DON'T go to the store.

But:
(i) it could equally well be argued that the correct response
would be {na ku go'i},
(ii) the desired interp of {na go'i} could equally well be handled
as part of the rules for interpreting {go'i}, e.g. if the rule
were that stuff in the {go'i} bridi has scope over the antecedent
of {go'i}, e.g.

A: xu do lo plise cu citka
B: go'i ca ro djedi
= "Every day I eat an apple"
not = "There is an apple that I eat every day"

Whereas in general the scope rules are very clear and simple, the
go'i interpretation rules are (afaik) a bit seat-of-the-pants. So
it seems like a bad idea to complicate the scope rules for the
sake of the go'i interp rules.

--And.

Jorge Llambias

unread,
Sep 28, 2002, 8:41:03 PM9/28/02
to loj...@yahoogroups.com

la pycyn cusku di'e

>English
>represents ~Fab by a~Fb, even when a is a quantified expression (though not
>always),

Maybe that holds sometimes for "all", but does it ever hold
for "some"? "Some guests are not here yet" is not ~ExFxb,
it is Ex~Fxb, I hope.

>so the way that natural langauges represent ~Fab is a~Fb (this is
>still pretty much true for SAE, with variants -- less true for French, if I
>remember rightly; I don't remember the distribution for Spanish cases;

In Spanish, saying "todos no" instead of "no todos" sounds
extremely weird to me. (Saying in English "Everybody doesn't"
instead of "Not everybody does" sounds just as weird to me, but
maybe it is because of the Spanish influence.) It can be done,
but in very restricted circumstances, with a very emphatic
"todos".

>pretty
>true for German), so Loglan (hence, Lojban) should represent ~Fab (even if
>a
>is quantified, i.e., ~ QxFxb) as a~Fb.

But Loglan kept it as ~ Qx xFb, didn't it?

>Yeah, I know, but that is sorta how
>it went and we are stuck with it (although I don't find it that hard to
>deal
>with, since it is so like (my) English).

It is not hard for simple sentences, but once you have two or three
quantifiers floating around, I find it very hard.

mu'o mi'e xorxes


_________________________________________________________________
Chat with friends online, try MSN Messenger: http://messenger.msn.com


------------------------ Yahoo! Groups Sponsor ---------------------~-->

jjllambias2000

unread,
Sep 27, 2002, 2:36:21 PM9/27/02
to loj...@yahoogroups.com

la jimc cusku di'e

> Perhaps the real issue is that you have to reprogram your semantic
analyser
> for real logic. Mapping Lojban 1-1 into an illogical natlang is
going to
> mangle the result, particularly where "carbon units" are most
sloppy in
> their logic.

Defining that {lo broda na brode} has to stand for {naku lo broda
cu brode} and not for {lo broda naku cu brode} is neither logical
nor illogical, it's just one possible convention. You can't say
that one expression is more logical than the other. Each has its
own logical meaning.

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->

jjllambias2000

unread,
Sep 27, 2002, 11:41:54 AM9/27/02
to loj...@yahoogroups.com

la djorden cusku di'e

> People also frequently forget boi when there's two lerfu sumti in
> a row, or ku when there's a joi for sumti.

Yes, but that seems to be a different type of error. Once you
understand the reason why the terminator is needed you don't
have to give a second thought to it. [I think numbers and lerfu
should not have been allowed to merge together, but that's for
some other day. Even so, the rule is learnable.] {na} is different.
Every time I see or write {na} in a longuish or a little bit complex
sentence I have to spend some time analyzing it before I can
be sure of what it says. And for the next sentence I have to
start from scratch, I haven't developed any intuition about it.

> I don't think the answer
> is language change, but rather more learning.

Maybe. I haven't given up on {na} yet, I do try to use it
by the Book, but I don't see myself making any progress with
it, as I do about other things. It may very well be that it
just takes longer to learn.

> (Most are still
> pretty much beginners to the language. Saying things wrong every
> now and then means we're broke, not the language).

Of course, and everybody says things wrong every now and then
even in their native language. All I'm saying is that the rule
for {na} is extremely difficult to master, at least for me.

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->

John Cowan

unread,
Sep 29, 2002, 12:14:59 AM9/29/02
to Jorge Llambias, loj...@yahoogroups.com
Jorge Llambias scripsit:

> In Spanish, saying "todos no" instead of "no todos" sounds
> extremely weird to me. (Saying in English "Everybody doesn't"
> instead of "Not everybody does" sounds just as weird to me, but
> maybe it is because of the Spanish influence.) It can be done,
> but in very restricted circumstances, with a very emphatic
> "todos".

Same story in English -- it is not the Spanish influence. Certain
words, especially those beginning with "every-", insist on attracting
the negation to them, so it has to be "No one writes to the Colonel"
and by no means "Everyone doesn't write to the Colonel" as a translation
of "El coronel no tiene quien le escriba" (the novel, that is).

--
Only do what only you can do. John Cowan <jco...@reutershealth.com>
--Edsger W. Dijkstra, http://www.reutershealth.com
deceased 6 August 2002 http://www.ccil.org/~cowan

------------------------ Yahoo! Groups Sponsor ---------------------~-->

Jordan DeLong

unread,
Sep 27, 2002, 11:54:02 AM9/27/02
to lojba...@lojban.org
On Fri, Sep 27, 2002 at 03:41:54PM -0000, jjllambias2000 wrote:
> la djorden cusku di'e
> > People also frequently forget boi when there's two lerfu sumti in
> > a row, or ku when there's a joi for sumti.
>
> Yes, but that seems to be a different type of error. Once you
> understand the reason why the terminator is needed you don't
> have to give a second thought to it. [I think numbers and lerfu
> should not have been allowed to merge together, but that's for
> some other day. Even so, the rule is learnable.] {na} is different.
> Every time I see or write {na} in a longuish or a little bit complex
> sentence I have to spend some time analyzing it before I can
> be sure of what it says. And for the next sentence I have to
> start from scratch, I haven't developed any intuition about it.
[...]

It's actually pretty simple: there's no need to do real thinking
about the sentence:
- if you can rephrase it as "It is false that: foo", the na is fine.

- if not, you can make it a "naku" and get position defined scope
- also consider a cmavo from NAhE, which is frequently better anyway.

This is easier for me than remembering {boi}! :)

--
Jordan DeLong - frac...@allusion.net
lu zo'o loi censa bakni cu terzba le zaltapla poi xagrai li'u
sei la mark. tuen. cusku

And Rosta

unread,
Sep 27, 2002, 11:59:32 AM9/27/02
to lojba...@lojban.org
Adam:

> 'na' can be interspersed among tenses, and I thought that this was to
> allow constructions such as 'roroi na' = 'always not', 'ka'e na' = 'is
> capable of not', etc. If NA must move to the front of the prenex, but
> tenses don't, then what is the point of having NA interspersed with
> tenses? Therefore, I think that tenses work the same as NA, with
> everything directly in front of the selbri moving to the left side of
> the prenex in the same order as they were before the selbri, followed
> by sumti and floating tenses in the order they appear in the bridi. So
> I think that the example from chapter 10 is wrong, since it ignores
> the different scope that tenses directly before the selbri have, and
> the example from chapter five is referring to the interaction between
> 'na' and 'pu', where there really is no change in meaning, but with
> tenses like 'roroi', you need to consider the order that the tenses
> and the na appear in.

>
> Does anyone object to this? Could anyone *really* think that 'roroi na
> broda' means 'not always brodas' (i.e. 'sometimes doesn't broda')? Can
> we get a (quasi-)official pronouncement from Cowan?

I sort of object to this. This is because I already object to the
rule for ku-less na, and in some ways I'd rather let it stand out
as egregiously exceptional, and not compound the problem by making
other tcita follow the same rules.

Instead, I'd suggest always using {na ku} and never plain {na}.
That or just ignore the special na-rule.

--And.

jjllambias2000

unread,
Sep 27, 2002, 11:51:58 AM9/27/02
to loj...@yahoogroups.com

la djan cusku di'e

> > Every time I see or write {na} in a longuish or a little bit

complex
> > sentence I have to spend some time analyzing it before I can
> > be sure of what it says. And for the next sentence I have to
> > start from scratch, I haven't developed any intuition about it.
>

> Whenever you see a sentence with "na", interpret it as "It is false
that ..."
> (or Spanish equivalent thereof).

Of course. That's what I meant by having to spend time analyzing it.
I have to reword it somehow and think of it in English or Spanish,
I can't just get the meaning directly from the Lojban text.

mu'o mi'e xorxes

------------------------ Yahoo! Groups Sponsor ---------------------~-->

Jordan DeLong

unread,
Sep 27, 2002, 8:07:03 PM9/27/02
to lojba...@lojban.org, John Cowan
On Sat, Sep 28, 2002 at 12:42:40AM +0100, And Rosta wrote:
> John:
> > And Rosta scripsit:
> > > To me it seems one of the more naturalistic features of Lojban --
> > > a quirky, exceptional, counterintuitive, unnecessary complication,
> > > of the sort natlangs are full of & Lojban is largely free of.
> >
> > What is alien is that a contradictory negation particle should be other
> > than at the beginning of the sentence.
> >
> > In Loglan, sentence-initial "no" served this function. IIRC, Lojbab
> > consciously moved it from the natural sentence-initial position to just
> > before the selbri "because it was more naturalistic". IMHO a mistake.
>
> As is obvious, I agree that the decision was a terrible mistake, but
> the idea that "it was more naturalistic" is fairly defensible,
> given that (a) some lects of English have it, and (b) quirkiness,
> exception-riddenness, counterintuiveness and unnecessary complication
> is highly characteristic of natlangs, as evidenced by the way that
> more accomplished naturalistic artlangers deliberately try to add
> it to their conlangs.

There's nothing to prevent you from always saying "naku" or always


putting it at the front of the bridi. (That is, unless the fact
that you never seem to produce any lojban text could be considered
preventative.)

--

Cyril Slobin

unread,
Sep 29, 2002, 9:24:41 PM9/29/02
to loj...@yahoogroups.com
On Sun, 29 Sep 2002, John Cowan wrote:

> Only do what only you can do.

Anyone to translate this in Lojban?

--
Cyril Slobin <slo...@ice.ru>

------------------------ Yahoo! Groups Sponsor ---------------------~-->

py...@aol.com

unread,
Sep 28, 2002, 6:42:35 PM9/28/02
to loj...@yahoogroups.com
In a message dated 9/27/2002 3:39:31 PM Central Daylight Time, jjlla...@hotmail.com writes:

<<
If I understand correctly, "the usual language of formal
logic" would have something like ~Fab

This can be described as:

1- Negation in front of the predicate
2- Negation in front of the whole expression

Lojban does: a~Fb, so as far as negation goes, it either follows
the usual language of formal logic (by 1) or it does not follow the
usual language of formal logic (by 2).

>>
Taxicab again.  Now I see your point: the Lojban is neither out in front marking out in front nor inside marking inside, so what is logical about it?  Answer -- historical and JCB-related (and we know what that means): English represents ~Fab by a~Fb, even when a is a quantified expression (though not always), so the way that natural langauges represent ~Fab is a~Fb (this is still pretty much true for SAE, with variants -- less true for French, if I remember rightly; I don't remember the distribution for Spanish cases; pretty true for German), so Loglan (hence, Lojban) should represent ~Fab (even if a is quantified, i.e., ~ QxFxb) as a~Fb.  Yeah, I know, but that is sorta how it went and we are stuck with it (although I don't find it that hard to deal with, since it is so like (my) English). 
To unsubscribe, send mail to lojban-un...@onelist.com

John Cowan

unread,
Sep 27, 2002, 11:45:24 AM9/27/02
to jjllambias2000, loj...@yahoogroups.com
jjllambias2000 scripsit:

> Every time I see or write {na} in a longuish or a little bit complex
> sentence I have to spend some time analyzing it before I can
> be sure of what it says. And for the next sentence I have to
> start from scratch, I haven't developed any intuition about it.

Whenever you see a sentence with "na", interpret it as "It is false that ..."
(or Spanish equivalent thereof).

> Of course, and everybody says things wrong every now and then


> even in their native language. All I'm saying is that the rule
> for {na} is extremely difficult to master, at least for me.

It is pretty alien, yes.

--
First known example of political correctness: John Cowan
"After Nurhachi had united all the other http://www.reutershealth.com
Jurchen tribes under the leadership of the http://www.ccil.org/~cowan
Manchus, his successor Abahai (1592-1643) jco...@reutershealth.com
issued an order that the name Jurchen should --S. Robert Ramsey,
be banned, and from then on, they were all _The Languages of China_
to be called Manchus.

Jordan DeLong

unread,
Sep 27, 2002, 11:31:00 AM9/27/02
to lojba...@lojban.org
On Fri, Sep 27, 2002 at 02:48:47PM -0000, jjllambias2000 wrote:
> la djan cusku di'e
>
> > The principle is that everything is exported
> > to the prenex in the order in which it (first) appears, *except* NA,
> > which is always exported to the very beginning. In that way,
> inserting
> > "na" before the selbri (mixed with tenses any way you like) is
> always the
> > exact contradictory negation of the version without "na".
>
> I doubt this can work in practice. {na} is consistently misused
> by almost everybody, so I suspect that the rule will be eventually
> generalized to order of appearance, {na} not excepted.

People also frequently forget boi when there's two lerfu sumti in
a row, or ku when there's a joi for sumti. I don't think the answer
is language change, but rather more learning. (Most are still


pretty much beginners to the language. Saying things wrong every
now and then means we're broke, not the language).

--

Jordan DeLong

unread,
Sep 27, 2002, 12:12:12 PM9/27/02
to lojba...@lojban.org
On Fri, Sep 27, 2002 at 03:51:58PM -0000, jjllambias2000 wrote:
> la djan cusku di'e
> > > Every time I see or write {na} in a longuish or a little bit
> complex
> > > sentence I have to spend some time analyzing it before I can
> > > be sure of what it says. And for the next sentence I have to
> > > start from scratch, I haven't developed any intuition about it.
> >
> > Whenever you see a sentence with "na", interpret it as "It is false
> that ..."
> > (or Spanish equivalent thereof).
>
> Of course. That's what I meant by having to spend time analyzing it.
> I have to reword it somehow and think of it in English or Spanish,
> I can't just get the meaning directly from the Lojban text.

So how about rephrasing into lojban instead of into english/spanish?
(le gerku na bajra == naku le gerku cu bajra)

py...@aol.com

unread,
Sep 27, 2002, 4:23:48 PM9/27/02
to loj...@yahoogroups.com
In a message dated 9/27/2002 1:37:41 PM Central Daylight Time, jjlla...@hotmail.com writes:

<<
Defining that {lo broda na brode} has to stand for {naku lo broda
cu brode} and not for {lo broda naku cu brode} is neither logical
nor illogical, it's just one possible convention. You can't say
that one expression is more logical than the other. Each has its
own logical meaning.

>>
As usual, "logical" applied to Lojban means "how to do it in the usual language of formal logic," which, in this case (as usually), is the standrad Lojban convention. (But there are perfectly good logical systems tht do it otherwise -- including especially ones that, like Lojban, are SVO rather than VSO.)
Yahoo! Groups Sponsor
ADVERTISEMENT

To unsubscribe, send mail to lojban-un...@onelist.com

Nora LeChevalier

unread,
Sep 27, 2002, 5:53:56 PM9/27/02
to lojba...@lojban.org
la xorxes cusku di'e:
>la djan cusku di'e
>
> > The principle is that everything is exported
> > to the prenex in the order in which it (first) appears, *except* NA,
> > which is always exported to the very beginning. In that way,
>inserting
> > "na" before the selbri (mixed with tenses any way you like) is
>always the
> > exact contradictory negation of the version without "na".
>
>I doubt this can work in practice. {na} is consistently misused
>by almost everybody, so I suspect that the rule will be eventually
>generalized to order of appearance, {na} not excepted.

"na" exporting to first in the prenex is necessary for "na go'i" to

work. "na go'i" MUST be the exact negation of the "xu" sentence to allow
answering questions to work.

xu do roroi klama le zarci
.i na go'i
You don't want "na go'i" to mean you always DON'T go to the store.


--
mi'e noras no...@lojban.org
Nora LeChevalier


Adam Raizen

unread,
Sep 26, 2002, 11:01:44 PM9/26/02
to lojba...@lojban.org
On IRC, a discussion of the interaction between tenses in front of the
selbri, floating tenses, and NA came up. I think that it is clear that
if we can agree on a way to move all of these to the prenex, we can
agree on their meaning. The Book, ch. 16, sec.9 (p.401) says:

"To represent a bridi negation using a prenex, remove the ``na'' from
before the selbri and place ``naku'' at the left end of the prenex.
This form is called ``external bridi negation'', as opposed to
``internal bridi negation'' using ``na''. The prenex version of
Example 9.1 is

"9.2) naku zo'u la djan. klama
It is not the case that: John comes.
It is false that: John comes."

On the other hand, ch. 10. sec. 13 (p. 234) says:

"13.5) puku mi ba klama le zarci
[past] I [future] go-to the market.
Earlier, I was going to go to the market.

"Here there are two tenses in the same bridi, the first floating free
and specified by ``puku'', the second in the usual place and specified
by ``ba''. They are considered cumulative in the same way as the two
tenses in separate sentences of Example 13.4. Example 13.5 is
therefore equivalent in meaning, except for emphasis, to:

"13.6) mi puba klama le zarci
I [past] [future] go-to the market.
I was going to go to the market."

and so it would seem that tenses are different, in that they get
placed in the prenex in the order in which they appear relative to
other tenses, whether or not they are directly in front of the selbri
or floating, but NA directly in front of the selbri always moves to
the left side of the prenex.

Finally, the book mentions the interaction between tenses and NA in
ch. 5, sec. 13 (p. 104):

"Various combinations of tense and bridi negation cmavo are permitted.
If both are expressed, either order is permissible with no change in
meaning:

"13.3) mi na pu klama le zarci
mi pu na klama le zarci
It is false that I went to the market.
I didn't go to the market."

If indeed tenses directly before the selbri do not go to the front of
the prenex along with NA, then constructions such 'roroi na broda' are


equivalent to 'na roroi broda', and both mean 'not always', rather
than the first having the obvious and logical meaning of 'always not',
i.e. 'never'.

'na' can be interspersed among tenses, and I thought that this was to
allow constructions such as 'roroi na' = 'always not', 'ka'e na' = 'is
capable of not', etc. If NA must move to the front of the prenex, but
tenses don't, then what is the point of having NA interspersed with
tenses? Therefore, I think that tenses work the same as NA, with
everything directly in front of the selbri moving to the left side of
the prenex in the same order as they were before the selbri, followed
by sumti and floating tenses in the order they appear in the bridi. So
I think that the example from chapter 10 is wrong, since it ignores
the different scope that tenses directly before the selbri have, and
the example from chapter five is referring to the interaction between
'na' and 'pu', where there really is no change in meaning, but with
tenses like 'roroi', you need to consider the order that the tenses
and the na appear in.

Does anyone object to this? Could anyone *really* think that 'roroi na
broda' means 'not always brodas' (i.e. 'sometimes doesn't broda')? Can
we get a (quasi-)official pronouncement from Cowan?

mu'o mi'e .adam.


Reply all
Reply to author
Forward
0 new messages