Re: [lojban] Re: possible worlds

1 view
Skip to first unread message

py...@aol.com

unread,
Jun 20, 2001, 9:03:44 PM6/20/01
to loj...@yahoogroups.com

--part1_c.17644141.2862a1f0_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/20/2001 6:25:42 PM Central Daylight Time,
r...@twcny.rr.com writes:


> .i le da'i logji cmavo poi roroi mapti zoi <if, then> cu du lo da'i logji
> cmavo poi da'inai na zasti
>
Something ain't quite right here -- and in a previous post from the same
source. I guess that -- whether discursive or "attitudinal" -- {da'i} can go
anywhere in a sentence, but it seems pretty clear that in at least some
places here it is adjectival to {logji cmavo}, meaning either "supposed" or
"{da'i}-like" ({se sruma}, more or less, or {me zo da'i} more or less). The
point is clear and certainly true, but the mode of expression leaves a bit to
be desired (am I about to be plonked again?)

--part1_c.17644141.2862a1f0_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/20/2001 6:25:42 PM Central Daylight Time,
<BR>r...@twcny.rr.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">.i le da'i logji cmavo poi roroi mapti zoi &lt;if, then&gt; cu du lo da'i logji
<BR>cmavo poi da'inai na zasti
<BR></BLOCKQUOTE>
<BR>Something ain't quite right here -- and in a previous post from the same
<BR>source. &nbsp;I guess that -- whether discursive or "attitudinal" -- {da'i} can go
<BR>anywhere in a sentence, but it seems pretty clear that in at least some
<BR>places here it is adjectival to {logji cmavo}, meaning either "supposed" or
<BR>"{da'i}-like" ({se sruma}, more or less, or {me zo da'i} more or less). &nbsp;The
<BR>point is clear and certainly true, but the mode of expression leaves a bit to
<BR>be desired (am I about to be plonked again?)</FONT></HTML>

Rob Speer

unread,
Jun 20, 2001, 3:20:49 PM6/20/01
to loj...@yahoogroups.com
la xod. pu cusku di'e
> ku'i pe'i lu ganai li'u goi ko'a roroi mapti

.i ni'i ma

> .i mi puza nalmulno ciksi secu'u
> http://groups.yahoo.com/group/lojban/message/7266

.i do djica tu'a le da'i logji cmavo poi na logji
.i ri na zasti

--
mu'o mi'e rab.spir


Bob LeChevalier (lojbab)

unread,
Jun 22, 2001, 10:07:33 AM6/22/01
to loj...@yahoogroups.com
At 11:03 AM 06/22/2001 +0000, Adam Raizen wrote:
>thin...@lycos.com wrote:
> > how about this? "if i were rich i'd have a house" goes to
> > "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> > UI cmavo to do this either, but i think there are substitutes.
>
>I can't bring myself to use "fau" for anything. Since "fasnu" only has
>one place, I have no idea what the relationship is supposed to be
>between things tagged with "fau" and the rest of the sentence.

What is the relationship between the event and the sentence in the English
phrase "In the event of blah, i will do blecch". There is clearly some
relationship between the event and the rest of the sentence, but it is
vague what that relationship is other than the event occurring is necessary.

Thus fau associates an event with a bridi and says that its occurrence is
in some way important to the truth of that bridi. Works rather like the
English.

lojbab
--
lojbab loj...@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273
Artificial language Loglan/Lojban: http://www.lojban.org


Invent Yourself

unread,
Jun 20, 2001, 1:54:25 PM6/20/01
to loj...@yahoogroups.com
On Wed, 20 Jun 2001, Robin Lee Powell wrote:

> On Tue, Jun 19, 2001 at 11:37:32PM -0400, Rob Speer wrote:
> > On Tue, Jun 19, 2001 at 10:56:14PM -0400, py...@aol.com wrote:
> > > In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,


> > > thin...@lycos.com writes:
> > >
> > >
> > > > how about this? "if i were rich i'd have a house" goes to
> > > > "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> > > > UI cmavo to do this either, but i think there are substitutes.
> > > >
>

> What's wrong with va'o?


dukse sucta .i ji'a le kamlogji cu melbi .i ji'a lu ganai li'u cu mapti le
sarcu gi'e na mapti le dukse

-----
We do not like And if a cat
those Rs and Ds, needed a hat?
Who can't resist Free enterprise
more subsidies. is there for that!


py...@aol.com

unread,
Jun 20, 2001, 6:03:03 PM6/20/01
to loj...@yahoogroups.com

--part1_113.94040a.28627797_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/20/2001 1:24:32 AM Central Daylight Time,
T...@fa-kuan.muc.de writes:


> *) It no longer seems to be a dead language, as Loglan obviously is present
> and vivid on the net with productive essays instead of
> whole surges of endless and boring threads most (interested!) people can no
> longer afford the time to follow.
>

Note that the paper is five years old. What has Loglan done lately?

--part1_113.94040a.28627797_boundary

Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/20/2001 1:24:32 AM Central Daylight Time,

<BR>T...@fa-kuan.muc.de writes:
<BR>
<BR>

<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">*) It no longer seems to be a dead language, as Loglan obviously is present
<BR>and vivid on the net with productive essays instead of
<BR>whole surges of endless and boring threads most (interested!) people can no
<BR>longer afford the time to follow.
<BR></BLOCKQUOTE>
<BR>
<BR>Note that the paper is five years old. &nbsp;What has Loglan done lately?</FONT></HTML>

Invent Yourself

unread,
Jun 20, 2001, 5:01:06 PM6/20/01
to loj...@yahoogroups.com
On Wed, 20 Jun 2001, Rob Speer wrote:

> la xod. pu cusku di'e
> > ku'i pe'i lu ganai li'u goi ko'a roroi mapti
>
> .i ni'i ma


do pu cusku le sedu'u su'oroi mapti .i tu'a mi natfe la'e di'u

> > .i mi puza nalmulno ciksi secu'u
> > http://groups.yahoo.com/group/lojban/message/7266
>
> .i do djica tu'a le da'i logji cmavo poi na logji
> .i ri na zasti

srana ma .i je'enai

Rob Speer

unread,
Jun 20, 2001, 1:40:41 AM6/20/01
to loj...@yahoogroups.com
.i la xod. pu cusku di'e
> .i pe'i lu ganai li'u na'o ralju le
> tadji be le nu fanva zoi zoi. if, then .zoi .iepei

.i ie
.i ga fu lu ganai broda gi brode li'u fanva fe zoi <if, then> ba'e ginai lejei
broda cu na se djuno
.i ganai
ge lejei broda cu se djuno ginai pilno zo da'i .a lo valsi poi simsa dy.
gi noda smuni gybygyby.

Rob Speer

unread,
Jun 19, 2001, 11:37:32 PM6/19/01
to loj...@yahoogroups.com
On Tue, Jun 19, 2001 at 10:56:14PM -0400, py...@aol.com wrote:
> In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,
> thin...@lycos.com writes:
>
>
> > how about this? "if i were rich i'd have a house" goes to
> > "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> > UI cmavo to do this either, but i think there are substitutes.
> >
>
> Nice, but what does it have to do with possible worlds? (the answer "nothing"
> is to its advantage) Is it clearly superior to {ganai mi ricfu gi mi ponse
> lo zdani}. The latter is, I suppose uninterestingly true, the former appears
> to be false or of uncertain -- and uncertainly obtainable -- truth value.
> So, as assertions, the advantage seems to lie with {ganai} form. But if this
> is something else than an assertion, it does seem to move over into the area
> of UI: intentions, hopes, and the like.

Eep. No. Even I, the vehement supporter of more widespread use of logical
connectives, say that you absolutely can't translate that with {ganai} or
anything of the sort. Since the statement presupposes that you are not rich,
the {ganai mi ricfu} part is necessarily true.

I thank xorxes for pointing out the existence of {da'i}; it seems to fit the
bill. However, what if you are describing someone else hypothesizing a possible
world? Do you use {da'idai} and get into two murky areas at once?

And do you use {da'i} with an ordinary logical connective, like:

{ganai da'i mi ricfu gi mi ponse lo zdani} ?

I still don't find this as elegant as Loglan's {foi} and {fio}; those would be
similar to Lojban modals or tenses, which would have a more useful role in the
grammar of the sentence than a UI cmavo. However, at least this doesn't seem to
leave a gaping conceptual hole in the language, and I suppose it's the best we
can realistically achieve within the baseline.
--
Rob Speer


Rob Speer

unread,
Jun 19, 2001, 6:28:00 PM6/19/01
to loj...@yahoogroups.com
On Tue, Jun 19, 2001 at 08:38:17PM -0000, A.W.T. wrote:
> Am I missing something in the Book or is the problem with possible worlds (i.e. the subjunctive etc. issue) still unsolved? Up to
> the moment I'm unsatisfied with sentences like "I should have done...", "If I were rich..." (Se fossi ricco mi comprarei una villa...)
> and the like.
> Are those venerable old gentlemen still sticking to the "dead language"* Loglan more aware of problems like those! Isn't e.g. the way
> pointed out at "http://www.loglan.org/Articles/I-would-if-I-could.html" a possible approach? IMHO, lojban attitudinals,
> evidentials, discursives etc. are very good and unique lojbanic features, yet do not solve here. The heavy discussion on extending
> their functional range (which I do not share) shows that there still is lack of things the kind mentioned above.

I have to agree with you here - I cannot quite understand how a UI cmavo
can switch us to another possible world in some contexts and leave us in the
same world in others. Though it's okay for attitudinals to be non-specific,
there needs to be a clear (and not subject to various interpretations) way of
saying such a sentence.

I think if we got this resolved, we could come to a useful conclusion on the
uses of the logical connectives - the last time that thread happened, it got
bogged down when the subjunctive (or Lojban's lack of it) reared its ugly head.

Incidentally, I'm not even sure which UI cmavo (besides .ei, which is an
attitudinal and thus cannot be used for anything specific) have been suggested
for this purpose, since I've always seen the subjunctive carefully avoided
instead of used.
--
Rob Speer


Edward Cherlin

unread,
Jun 24, 2001, 5:19:38 PM6/24/01
to loj...@yahoogroups.com
At 01:23 PM 6/23/2001, Adam Raizen wrote:
>la lojbab cusku di'e

>
> > What is the relationship between the event and the sentence in the
>English
> > phrase "In the event of blah, i will do blecch". There is clearly
>some
> > relationship between the event and the rest of the sentence, but it
>is
> > vague what that relationship is other than the event occurring is
>necessary.
> >
> > Thus fau associates an event with a bridi and says that its
>occurrence is
> > in some way important to the truth of that bridi. Works rather like
>the
> > English.
>
>At least with the English, the idiom *does* mean at least that if the
>event occurs, the main sentence also occurs.

This is ambiguous, but it is commonly a statement of intention, not of fact.

We do not have a clear modal logic model in Lojban for the meanings of such
statements and the evaluaton of their truth. Modal logics study statements
about possibility and necessity, probability, intention, permission,
belief, and the like. I can dig out an old textbook and post some notes, if
anybody thinks it will help.

We also don't have any useful logical model of *impossible* worlds and
other weird ontologies. I don't know whether anyone has successfully
constructed one, but I could look. Quine wrote about these problems, such
as the "square circle" and "Plato's beard" but certainly didn't resolve them.

py...@aol.com

unread,
Jun 21, 2001, 12:04:43 PM6/21/01
to loj...@yahoogroups.com

--part1_e5.80a6d99.2863751b_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/20/2001 9:40:10 PM Central Daylight Time,
r...@twcny.rr.com writes:


> On Wed, Jun 20, 2001 at 09:03:44PM -0400, py...@aol.com wrote:
> > In a message dated 6/20/2001 6:25:42 PM Central Daylight Time,
> > r...@twcny.rr.com writes:
> >
> >

> > > .i le da'i logji cmavo poi roroi mapti zoi <if, then> cu du lo da'i
> logji


> > > cmavo poi da'inai na zasti
> > >
> > Something ain't quite right here -- and in a previous post from the same
> > source. I guess that -- whether discursive or "attitudinal" -- {da'i}
> can go
> > anywhere in a sentence, but it seems pretty clear that in at least some
> > places here it is adjectival to {logji cmavo}, meaning either "supposed"
> or
> > "{da'i}-like"
>

> Why would it be adjectival? If I were talking about the word {da'i} itself,
> I
> would have said something involving {zo da'i}. Here I was using it for its
> newfound purpose of describing possible worlds.
>
> What I was using it for was "the supposed logical connective which always
> applies to <if, then>".
>
> Without the {da'i}, I would be talking about "the logical connective which
> always applies to <if, then>". However, no such connective exists, and that
> sentence would logically fall apart because of that. So I used {da'i} to
> refer to this object in a possible world where such a thing would exist (and
> I pity the inhabitants of that world and the broken version of Lojban
> they're
> stuck with).
>

The point is well taken, as I said, but will this way of saying it really
work? Wherever {da'i} occurs, it presumably works to throw the whole
sentence into the suppositive mood (I'm using the official rules, of
course). Whether the repeated {da'i} throws it into a second-order
supposition or not, I can't say, nor can I work out the rhetorical effect of
putting the {da'i} after {le}. At a guess the latter focuses the goal of the
supposition on the sumti which {le} begins, which is, I think, your goal,
more or less.
So this seems to say "Suppose that there is a logical connective which always
matches 'if then' ..." or, more literally but clearly not what you want,
"Suppose that the logical connective which always matches 'if then' is a
logical connective that does not in fact exist" Now all of this does make
for a problem, since it involves a referring phrase which you want to say
does not refer. And yet it does refer (in fact, to {ganai...gi...}); what it
does not do is match "if..., then..."
Does {le a'o mi se prami} mean "the beloved I hope for"?

--part1_e5.80a6d99.2863751b_boundary

Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/20/2001 9:40:10 PM Central Daylight Time,

<BR>r...@twcny.rr.com writes:
<BR>
<BR>

<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Wed, Jun 20, 2001 at 09:03:44PM -0400, py...@aol.com wrote:
<BR>&gt; In a message dated 6/20/2001 6:25:42 PM Central Daylight Time,
<BR>&gt; r...@twcny.rr.com writes:
<BR>&gt;
<BR>&gt;
<BR>&gt; &gt; .i le da'i logji cmavo poi roroi mapti zoi &lt;if, then&gt; cu du lo da'i
<BR>logji
<BR>&gt; &gt; cmavo poi da'inai na zasti
<BR>&gt; &gt;
<BR>&gt; Something ain't quite right here -- and in a previous post from the same
<BR>&gt; source. &nbsp;I guess that -- whether discursive or "attitudinal" -- {da'i}
<BR>can go
<BR>&gt; anywhere in a sentence, but it seems pretty clear that in at least some
<BR>&gt; places here it is adjectival to {logji cmavo}, meaning either "supposed"
<BR>or
<BR>&gt; "{da'i}-like"
<BR>
<BR>Why would it be adjectival? If I were talking about the word {da'i} itself,
<BR>I
<BR>would have said something involving {zo da'i}. Here I was using it for its
<BR>newfound purpose of describing possible worlds.
<BR>
<BR>What I was using it for was "the supposed logical connective which always
<BR>applies to &lt;if, then&gt;".
<BR>
<BR>Without the {da'i}, I would be talking about "the logical connective which
<BR>always applies to &lt;if, then&gt;". However, no such connective exists, and that
<BR>sentence would logically fall apart because of that. So I used {da'i} to
<BR>refer to this object in a possible world where such a thing would exist (and
<BR>I pity the inhabitants of that world and the broken version of Lojban
<BR>they're
<BR>stuck with).
<BR></BLOCKQUOTE>
<BR>
<BR>The point is well taken, as I said, but will this way of saying it really
<BR>work? &nbsp;Wherever {da'i} occurs, it presumably works to throw the whole
<BR>sentence into the suppositive mood &nbsp;(I'm using the official rules, of
<BR>course). &nbsp;Whether the repeated {da'i} throws it into a second-order
<BR>supposition or not, I can't say, nor can I work out the rhetorical effect of
<BR>putting the {da'i} after {le}. &nbsp;At a guess the latter focuses the goal of the
<BR>supposition on the sumti which {le} begins, which is, I think, your goal,
<BR>more or less.
<BR>So this seems to say "Suppose that there is a logical connective which always
<BR>matches 'if then' ..." or, more literally but clearly not what you want,
<BR>"Suppose that the logical connective which always matches 'if then' is a
<BR>logical connective that does not in fact exist" &nbsp;Now all of this does make
<BR>for a problem, since it involves a referring phrase which you want to say
<BR>does not refer. And yet it does refer (in fact, to {ganai...gi...}); what it
<BR>does not do is match "if..., then..."
<BR>Does {le a'o mi se prami} mean "the beloved I hope for"?</FONT></HTML>

Rob Speer

unread,
Jun 20, 2001, 3:24:09 PM6/20/01
to loj...@yahoogroups.com
On Wed, Jun 20, 2001 at 10:21:46AM -0700, Robin Lee Powell wrote:
> > I still don't find this as elegant as Loglan's {foi} and {fio}; those would be
> > similar to Lojban modals or tenses, which would have a more useful role in the
> > grammar of the sentence than a UI cmavo. However, at least this doesn't seem to
> > leave a gaping conceptual hole in the language, and I suppose it's the best we
> > can realistically achieve within the baseline.
>
> What's wrong with va'o?
>
> -Robin

Nobody knows what it really means, for one thing.

Also, most proponents of va'o suggest using it for ordinary if...then, which
means it can't be a possible-world indicator as well.

If va'o could be taken over to solely refer to possible worlds (and ordinary
logic would use [gasp] the logical connectives!) it would be great, but I can't
see that happening.
--
Rob Speer


Rob Speer

unread,
Jun 20, 2001, 7:17:12 PM6/20/01
to loj...@yahoogroups.com
la xod. pu cusku di'e
> On Wed, 20 Jun 2001, Rob Speer wrote:
>
> > la xod. pu cusku di'e
> > > ku'i pe'i lu ganai li'u goi ko'a roroi mapti
> >
> > .i ni'i ma
>
>
> do pu cusku le sedu'u su'oroi mapti .i tu'a mi natfe la'e di'u

.i go'i fi ma
.i do darlu fi le logji ki'ulenu ko'a na roroi mapti
.i ke'u go ko'a mapti gi lejei broda kei na se djuno

> > > .i mi puza nalmulno ciksi secu'u
> > > http://groups.yahoo.com/group/lojban/message/7266
> >
> > .i do djica tu'a le da'i logji cmavo poi na logji
> > .i ri na zasti
>
> srana ma .i je'enai

.i le da'i logji cmavo poi roroi mapti zoi <if, then> cu du lo da'i logji
cmavo poi da'inai na zasti

py...@aol.com

unread,
Jun 25, 2001, 4:32:08 AM6/25/01
to loj...@yahoogroups.com

--part1_112.c0ff77.28685108_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/24/2001 4:13:08 PM Central Daylight Time,
edward.che...@aya.yale.edu writes:


> We do not have a clear modal logic model in Lojban for the meanings of such
> statements and the evaluaton of their truth. Modal logics study statements
> about possibility and necessity, probability, intention, permission,
> belief, and the like. I can dig out an old textbook and post some notes, if
> anybody thinks it will help.
>
> We also don't have any useful logical model of *impossible* worlds and
> other weird ontologies. I don't know whether anyone has successfully
> constructed one, but I could look. Quine wrote about these problems, such
> as the "square circle" and "Plato's beard" but certainly didn't resolve
>

Probably don't have them because logicians have tended to multiply the
possibilities without coming down very hard for one of them as right. Prior
always used to insist that it was the logicians' job to do just that, leaving
it to experts in whatever to decide what was right. But there are precious
few experts in most of the areas that modal logic encompasses -- indeed, few
recognized areas, even.
Lojban has gotten along with an informal notion of "possible worlds" -- which
I now think may be a mistake -- and an even less clear notion of a situation
(which allows some work, at least, with impossible worlds). I am inclined --
in keeping with the almost concurrent thread -- to see the need for more
inquiry into the purposes for which language is used. As Cherlin points out,
a lot of contrary-to-fact comments are intentions, not descriptions. Others
are simply disguised generalizations (whatever those are). Others may be
more like hopes. But we don't seem to know much about what the range of
possibilities is here yet, nor how they are characterized, nor how they are
represented in languages already fully developed. I think there is a good
deal of work in the area of Chapters 13 and 14 (and, I suppose, 9, 10 and 11)
before we have a reasonably satisfactory understanding. Oh and serious
thoughts about what happens in English and other languages we live in as well.


--part1_112.c0ff77.28685108_boundary

Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/24/2001 4:13:08 PM Central Daylight Time,

<BR>edward.che...@aya.yale.edu writes:
<BR>
<BR>

<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">We do not have a clear modal logic model in Lojban for the meanings of such
<BR>statements and the evaluaton of their truth. Modal logics study statements
<BR>about possibility and necessity, probability, intention, permission,
<BR>belief, and the like. I can dig out an old textbook and post some notes, if
<BR>anybody thinks it will help.
<BR>
<BR>We also don't have any useful logical model of *impossible* worlds and
<BR>other weird ontologies. I don't know whether anyone has successfully
<BR>constructed one, but I could look. Quine wrote about these problems, such
<BR>as the "square circle" and "Plato's beard" but certainly didn't resolve
<BR>them.</BLOCKQUOTE>
<BR>
<BR>Probably don't have them because logicians have tended to multiply the
<BR>possibilities without coming down very hard for one of them as right. &nbsp;Prior
<BR>always used to insist that it was the logicians' job to do just that, leaving
<BR>it to experts in whatever to decide what was right. &nbsp;But there are precious
<BR>few experts in most of the areas that modal logic encompasses -- indeed, few
<BR>recognized areas, even. &nbsp;
<BR>Lojban has gotten along with an informal notion of "possible worlds" -- which
<BR>I now think may be a mistake -- and an even less clear notion of a situation
<BR>(which allows some work, at least, with impossible worlds). &nbsp;I am inclined --
<BR>in keeping with the almost concurrent thread -- to see the need for more
<BR>inquiry into the purposes for which language is used. &nbsp;As Cherlin points out,
<BR>a lot of contrary-to-fact comments are intentions, not descriptions. &nbsp;Others
<BR>are simply disguised generalizations (whatever those are). &nbsp;Others may be
<BR>more like hopes. &nbsp;But we don't seem to know much about what the range of
<BR>possibilities is here yet, nor how they are characterized, nor how they are
<BR>represented in languages already fully developed. I think there is a good
<BR>deal of work in the area of Chapters 13 and 14 (and, I suppose, 9, 10 and 11)
<BR>before we have a reasonably satisfactory understanding. &nbsp;Oh and serious
<BR>thoughts about what happens in English and other languages we live in as well.
<BR></FONT></HTML>

A.W.T.

unread,
Jun 19, 2001, 4:38:17 PM6/19/01
to loj...@yahoogroups.com
Am I missing something in the Book or is the problem with possible worlds (i.e. the subjunctive etc. issue) still unsolved? Up to
the moment I'm unsatisfied with sentences like "I should have done...", "If I were rich..." (Se fossi ricco mi comprarei una villa...)
and the like.
Are those venerable old gentlemen still sticking to the "dead language"* Loglan more aware of problems like those! Isn't e.g. the way
pointed out at "http://www.loglan.org/Articles/I-would-if-I-could.html" a possible approach? IMHO, lojban attitudinals,
evidentials, discursives etc. are very good and unique lojbanic features, yet do not solve here. The heavy discussion on extending
their functional range (which I do not share) shows that there still is lack of things the kind mentioned above.

*) It no longer seems to be a dead language, as Loglan obviously is present and vivid on the net with productive essays instead of

whole surges of endless and boring threads most (interested!) people can no longer afford the time to follow.

mi'e .aulun.

Rob Speer

unread,
Jun 20, 2001, 10:30:47 PM6/20/01
to loj...@yahoogroups.com
On Wed, Jun 20, 2001 at 09:03:44PM -0400, py...@aol.com wrote:
> In a message dated 6/20/2001 6:25:42 PM Central Daylight Time,
> r...@twcny.rr.com writes:
>
>
> > .i le da'i logji cmavo poi roroi mapti zoi <if, then> cu du lo da'i logji
> > cmavo poi da'inai na zasti
> >
> Something ain't quite right here -- and in a previous post from the same
> source. I guess that -- whether discursive or "attitudinal" -- {da'i} can go
> anywhere in a sentence, but it seems pretty clear that in at least some
> places here it is adjectival to {logji cmavo}, meaning either "supposed" or
> "{da'i}-like"

Why would it be adjectival? If I were talking about the word {da'i} itself, I


would have said something involving {zo da'i}. Here I was using it for its
newfound purpose of describing possible worlds.

What I was using it for was "the supposed logical connective which always
applies to <if, then>".

Without the {da'i}, I would be talking about "the logical connective which
always applies to <if, then>". However, no such connective exists, and that
sentence would logically fall apart because of that. So I used {da'i} to
refer to this object in a possible world where such a thing would exist (and
I pity the inhabitants of that world and the broken version of Lojban they're
stuck with).

--
Rob Speer


michael helsem

unread,
Jun 25, 2001, 11:25:18 AM6/25/01
to loj...@yahoogroups.com
>From: Edward Cherlin <edward.che...@aya.yale.edu>
li'o

>We also don't have any useful logical model of *impossible* worlds and
>other weird ontologies.

not DA'I or PE'A? why must it require an ontology? i think we
keep making too much of this. such statements in the vernacular
aren't all that complex!
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com


Invent Yourself

unread,
Jun 19, 2001, 11:59:51 PM6/19/01
to loj...@yahoogroups.com
On Tue, 19 Jun 2001, Rob Speer wrote:

> On Tue, Jun 19, 2001 at 10:56:14PM -0400, py...@aol.com wrote:
> > In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,
> > thin...@lycos.com writes:
> >
> >
> > > how about this? "if i were rich i'd have a house" goes to
> > > "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> > > UI cmavo to do this either, but i think there are substitutes.
> > >
> >
> > Nice, but what does it have to do with possible worlds? (the answer "nothing"
> > is to its advantage) Is it clearly superior to {ganai mi ricfu gi mi ponse
> > lo zdani}. The latter is, I suppose uninterestingly true, the former appears
> > to be false or of uncertain -- and uncertainly obtainable -- truth value.
> > So, as assertions, the advantage seems to lie with {ganai} form. But if this
> > is something else than an assertion, it does seem to move over into the area
> > of UI: intentions, hopes, and the like.
>
> Eep. No. Even I, the vehement supporter of more widespread use of logical
> connectives, say that you absolutely can't translate that with {ganai} or
> anything of the sort. Since the statement presupposes that you are not rich,
> the {ganai mi ricfu} part is necessarily true.

u'i .i ga'inai .e'o ko skicu le krinu be le si'o lu ganai li'u cu xlali

> I thank xorxes for pointing out the existence of {da'i}; it seems to fit the
> bill. However, what if you are describing someone else hypothesizing a possible
> world? Do you use {da'idai} and get into two murky areas at once?

da'ipei

> And do you use {da'i} with an ordinary logical connective, like:
>
> {ganai da'i mi ricfu gi mi ponse lo zdani} ?


.e'a

Adam Raizen

unread,
Jun 23, 2001, 4:23:19 PM6/23/01
to loj...@yahoogroups.com
la lojbab cusku di'e

> What is the relationship between the event and the sentence in the


English
> phrase "In the event of blah, i will do blecch". There is clearly
some
> relationship between the event and the rest of the sentence, but it
is
> vague what that relationship is other than the event occurring is
necessary.
>
> Thus fau associates an event with a bridi and says that its
occurrence is
> in some way important to the truth of that bridi. Works rather like
the
> English.

At least with the English, the idiom *does* mean at least that if the
event occurs, the main sentence also occurs. I don't think it's so
clear that the relationship in Lojban has something to do with the
event occurring being necessary to the main bridi. To me, "fau le nu
mi ricfu kei mi ponse lo zdani" means something like "I have a house
with my being rich being an event." With "va'o" and ilk, at least,
there are two places, and we can argue and define the relationship
between them as much as we want, but with "fau", there's just nothing
to work with.

I realize that the BAI cmavo were made before the parallelism with the
place structures of the gismu was formalized, but now "fau" seems to
be a rather useless word.

mu'o mi'e adam

Robin Lee Powell

unread,
Jun 20, 2001, 1:21:46 PM6/20/01
to loj...@yahoogroups.com
On Tue, Jun 19, 2001 at 11:37:32PM -0400, Rob Speer wrote:
> On Tue, Jun 19, 2001 at 10:56:14PM -0400, py...@aol.com wrote:
> > In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,
> > thin...@lycos.com writes:
> >
> >
> > > how about this? "if i were rich i'd have a house" goes to
> > > "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> > > UI cmavo to do this either, but i think there are substitutes.
> > >
> >
> > Nice, but what does it have to do with possible worlds? (the answer "nothing"
> > is to its advantage) Is it clearly superior to {ganai mi ricfu gi mi ponse
> > lo zdani}. The latter is, I suppose uninterestingly true, the former appears
> > to be false or of uncertain -- and uncertainly obtainable -- truth value.
> > So, as assertions, the advantage seems to lie with {ganai} form. But if this
> > is something else than an assertion, it does seem to move over into the area
> > of UI: intentions, hopes, and the like.

[snip]

> I still don't find this as elegant as Loglan's {foi} and {fio}; those would be
> similar to Lojban modals or tenses, which would have a more useful role in the
> grammar of the sentence than a UI cmavo. However, at least this doesn't seem to
> leave a gaping conceptual hole in the language, and I suppose it's the best we
> can realistically achieve within the baseline.

What's wrong with va'o?

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ BTW, I'm male, honest.
le datni cu djica le nu zifre .iku'i .oi le so'e datni cu to'e te pilno
je xlali -- RLP http://www.lojban.org/

Rob Speer

unread,
Jun 20, 2001, 12:52:40 AM6/20/01
to loj...@yahoogroups.com
.i la kokofint. zo'o pu cusku di'e

> > Eep. No. Even I, the vehement supporter of more widespread use of logical
> > connectives, say that you absolutely can't translate that with {ganai} or
> > anything of the sort. Since the statement presupposes that you are not rich,
> > the {ganai mi ricfu} part is necessarily true.
>
>
>
> u'i .i ga'inai .e'o ko skicu le krinu be le si'o lu ganai li'u cu xlali

.i pukiku le notci befi la xorxes. cu tolcliva calenu mi ciska de'u
.i mi galfi jinvi fi lenu ta'i ma cusku .ei tu'a le cumki munje
.i mi galfi lemi notci .iku'i mi na galfi le pamoi selju'a

.i nau pe'i xamgu fa lenu pilno lu ganaida'i broda gi brode li'u ki'u lenu
claxu lo te cuxna

Invent Yourself

unread,
Jun 20, 2001, 1:26:49 PM6/20/01
to loj...@yahoogroups.com

ku'i pe'i lu ganai li'u goi ko'a roroi mapti

.i mi puza nalmulno ciksi secu'u
http://groups.yahoo.com/group/lojban/message/7266

-----

Rob Speer

unread,
Jun 20, 2001, 3:30:55 PM6/20/01
to loj...@yahoogroups.com
On Wed, Jun 20, 2001 at 07:19:38PM +0000, michael helsem wrote:
> >From: "Jorge Llambias"
> li'o
> >Is this an emotional topic or what?
> >
>
> da'i go'i
("It supposedly is [in some possible world].")

.i ji'a da'inai go'e
--
Rob Speer


Jorge Llambias

unread,
Jun 19, 2001, 8:07:21 PM6/19/01
to loj...@yahoogroups.com

la rab cusku di'e

>Incidentally, I'm not even sure which UI cmavo (besides .ei, which is an
>attitudinal and thus cannot be used for anything specific) have been
>suggested
>for this purpose, since I've always seen the subjunctive carefully avoided
>instead of used.

The main one for counterfactuals is {da'i} (and it _has_ been used).
But it is hard to get to it when people get so ruffled about much
more basic stuff. Is this an emotional topic or what?

mu'o mi'e xorxes


_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.


Adam Raizen

unread,
Jun 22, 2001, 7:03:26 AM6/22/01
to loj...@yahoogroups.com
thin...@lycos.com wrote:

> how about this? "if i were rich i'd have a house" goes to
> "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> UI cmavo to do this either, but i think there are substitutes.

I can't bring myself to use "fau" for anything. Since "fasnu" only has


one place, I have no idea what the relationship is supposed to be

between things tagged with "fau" and the rest of the sentence.

Jorge Llambias

unread,
Jun 21, 2001, 9:31:47 PM6/21/01
to loj...@yahoogroups.com

la pycyn cusku di'e

> > > > .i le da'i logji cmavo poi roroi mapti zoi <if, then>
> > > > cu du lo da'i logji cmavo poi da'inai na zasti
>

>or, more literally but clearly not what you want,
>"Suppose that the logical connective which always matches 'if then' is a
>logical connective that does not in fact exist"

Maybe:

i le logji cmavo poi da'i roroi mapti zoi <if, then>
cu du lo logji cmavo poi da'inai na zasti

which at least has {da'i} marking a whole subclause bridi.

>Does {le a'o mi se prami} mean "the beloved I hope for"?

I'm not sure. It is tempting though...

lo <UI> broda = da poi <UI> broda

It might even work...

Rob Speer

unread,
Jun 21, 2001, 2:55:37 PM6/21/01
to loj...@yahoogroups.com
On Thu, Jun 21, 2001 at 12:04:43PM -0400, py...@aol.com wrote:
> The point is well taken, as I said, but will this way of saying it really
> work? Wherever {da'i} occurs, it presumably works to throw the whole
> sentence into the suppositive mood (I'm using the official rules, of
> course). Whether the repeated {da'i} throws it into a second-order
> supposition or not, I can't say, nor can I work out the rhetorical effect of
> putting the {da'i} after {le}. At a guess the latter focuses the goal of the
> supposition on the sumti which {le} begins, which is, I think, your goal,
> more or less.
> So this seems to say "Suppose that there is a logical connective which always
> matches 'if then' ..." or, more literally but clearly not what you want,
> "Suppose that the logical connective which always matches 'if then' is a
> logical connective that does not in fact exist" Now all of this does make
> for a problem, since it involves a referring phrase which you want to say
> does not refer. And yet it does refer (in fact, to {ganai...gi...}); what it
> does not do is match "if..., then..."
> Does {le a'o mi se prami} mean "the beloved I hope for"?

I suppose you're right. So now we're back where we were, because the "possible
worlds" cmavo can't be a UI - it doesn't have enough grammatical structure that
way.

It may have to be a xVV cmavo. (It feels icky to use xVV cmavo when there are
empty cmavo like {bi'a} and {ci'a} at our disposal, but oh well.) If so, I
would suggest that it should be a tense.
--
Rob Speer


thin...@lycos.com

unread,
Jun 19, 2001, 7:40:22 PM6/19/01
to loj...@yahoogroups.com

how about this? "if i were rich i'd have a house" goes to

py...@aol.com

unread,
Jun 21, 2001, 10:52:37 PM6/21/01
to loj...@yahoogroups.com

--part1_7e.16bdbc03.28640cf5_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/21/2001 7:42:56 PM Central Daylight Time,
r...@twcny.rr.com writes:


> I suppose you're right. So now we're back where we were, because the
> "possible
> worlds" cmavo can't be a UI - it doesn't have enough grammatical structure
> that
> way.
>
> It may have to be a xVV cmavo. (It feels icky to use xVV cmavo when there
> are
> empty cmavo like {bi'a} and {ci'a} at our disposal, but oh well.) If so, I
> would suggest that it should be a tense.
>

Oh, the grammar is not all that bad. We need, however, to decide what the
critter is to do. As I have said, I think the "possible worlds" talk --
outside of technical logic -- is a pretty bad one, since it goes so far
toward reducing all speech acts to describing/asserting. In that view,
though, making moves to other worlds would be rather like a tense, except
that the world would be described situationally only, never by displacement
(what would be the metric of displacement after all -- or even the
direction?) I think that the UIs work rather better: they describe the
displacement and permit laying stress when that is useful (it's the change in
my economic status, not a change in me that is important in {mi ricfu da'i},
say). Which of these words we will use ({da'i}, {va'o} in a different way,
probably some others) and how they correlate with various brivla (I think
{sruma} goes nicely with {da'i} despite the different sources), is going to
depend upon what we find when we really start looking at the various kinds of
acts that may be out there: telling a tale is different from doing a proof
and that from contingency plannig or speculation about character or... . We
need to find out what our resources are as well (what words can reasonably
play a role here?)
As for your sentence, I think all you really needed to say was {ledo logji
cmavo poi roroi mapti zoi <if, then> na da'inai roroi mapti zoi <if, then>}
"What you call the logical connective that always matches "if, then" does not
in fact always match "if, then":" the putative conditional doesn't work".
Otherwise, I'd stick {se sruma} in.

--part1_7e.16bdbc03.28640cf5_boundary

Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/21/2001 7:42:56 PM Central Daylight Time,
<BR>r...@twcny.rr.com writes:
<BR>
<BR>
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">I suppose you're right. So now we're back where we were, because the
<BR>"possible
<BR>worlds" cmavo can't be a UI - it doesn't have enough grammatical structure
<BR>that
<BR>way.
<BR>
<BR>It may have to be a xVV cmavo. (It feels icky to use xVV cmavo when there
<BR>are
<BR>empty cmavo like {bi'a} and {ci'a} at our disposal, but oh well.) If so, I
<BR>would suggest that it should be a tense.
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>Oh, the grammar is not all that bad. &nbsp;We need, however, to decide what the
<BR>critter is to do. &nbsp;As I have said, I think the "possible worlds" talk --
<BR>outside of technical logic -- is a pretty bad one, since it goes so far
<BR>toward reducing all speech acts to describing/asserting. &nbsp;In that view,
<BR>though, making moves to other worlds would be rather like a tense, except
<BR>that the world would be described situationally only, never by displacement
<BR>(what would be the metric of displacement after all -- or even the
<BR>direction?) &nbsp;I think that the UIs work rather better: they describe the
<BR>displacement and permit laying stress when that is useful (it's the change in
<BR>my economic status, not a change in me that is important in {mi ricfu da'i},
<BR>say). &nbsp;Which of these words we will use ({da'i}, {va'o} in a different way,
<BR>probably some others) and how they correlate with various brivla (I think
<BR>{sruma} goes nicely with {da'i} despite the different sources), is going to
<BR>depend upon what we find when we really start looking at the various kinds of
<BR>acts that may be out there: telling a tale is different from doing a proof
<BR>and that from contingency plannig or speculation about character or... . &nbsp;We
<BR>need to find out what our resources are as well (what words can reasonably
<BR>play a role here?)
<BR>As for your sentence, I think all you really needed to say was {ledo logji
<BR>cmavo poi roroi mapti zoi &lt;if, then&gt; na da'inai roroi mapti zoi &lt;if, then&gt;} &nbsp;
<BR>"What you call the logical connective that always matches "if, then" does not
<BR>in fact always match "if, then":" the putative conditional doesn't work". &nbsp;
<BR>Otherwise, I'd stick {se sruma} in.</FONT></HTML>

Bob LeChevalier (lojbab)

unread,
Jun 20, 2001, 1:29:17 AM6/20/01
to loj...@yahoogroups.com
At 08:38 PM 06/19/2001 +0000, A.W.T. wrote:
>*) It no longer seems to be a dead language, as Loglan obviously is
>present and vivid on the net with productive essays instead of
>whole surges of endless and boring threads most (interested!) people can
>no longer afford the time to follow.

It is? I haven't seen anything. The TLI Loglan site is ancient, still
announcing JCB's death as "news". It was last updated a year ago, but I
think that was when they moved to the loglan.org site, and there isn't any
new content.

The Russian Loglanist may indeed be doing something on his web site. I
haven't looked.

The TLI mailing list has not had any postings at all since the end of April
(except for 1 piece of spam of the Nigerian oil scam).

The "productive essay" you cited was written back in 1996, and we produced
a 600 page book after that.

lojbab

ps. If you know of any new TLI Loglan work, I'd be interested in hearing
about it.
--
lojbab loj...@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273
Artificial language Loglan/Lojban: http://www.lojban.org


py...@aol.com

unread,
Jun 19, 2001, 10:56:14 PM6/19/01
to loj...@yahoogroups.com

--part1_8e.17392b9c.28616ace_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,
thin...@lycos.com writes:


> how about this? "if i were rich i'd have a house" goes to
> "fau le nu mi ricfu kei mi ponse lo zdani". i don't like using the
> UI cmavo to do this either, but i think there are substitutes.
>

Nice, but what does it have to do with possible worlds? (the answer "nothing"

is to its advantage) Is it clearly superior to {ganai mi ricfu gi mi ponse
lo zdani}. The latter is, I suppose uninterestingly true, the former appears
to be false or of uncertain -- and uncertainly obtainable -- truth value.
So, as assertions, the advantage seems to lie with {ganai} form. But if this
is something else than an assertion, it does seem to move over into the area
of UI: intentions, hopes, and the like.

--part1_8e.17392b9c.28616ace_boundary

Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><BODY BGCOLOR="#ffffff"><FONT SIZE=2>In a message dated 6/19/2001 8:10:08 PM Central Daylight Time,

<BR>thin...@lycos.com writes:
<BR>
<BR>

<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">how about this? &nbsp;"if i were rich i'd have a house" goes to
<BR>"fau le nu mi ricfu kei mi ponse lo zdani". &nbsp;i don't like using the
<BR>UI cmavo to do this either, but i think there are substitutes.
<BR></BLOCKQUOTE>
<BR>
<BR>Nice, but what does it have to do with possible worlds? (the answer "nothing"
<BR>is to its advantage) &nbsp;Is it clearly superior to {ganai mi ricfu gi mi ponse
<BR>lo zdani}. &nbsp;The latter is, I suppose uninterestingly true, the former appears
<BR>to be false or of uncertain -- and uncertainly obtainable -- truth value. &nbsp;
<BR>So, as assertions, the advantage seems to lie with {ganai} form. &nbsp;But if this
<BR>is something else than an assertion, it does seem to move over into the area
<BR>of UI: intentions, hopes, and the like.</FONT></HTML>

michael helsem

unread,
Jun 20, 2001, 3:19:38 PM6/20/01
to loj...@yahoogroups.com
>From: "Jorge Llambias"
li'o
>Is this an emotional topic or what?
>

da'i go'i

thin...@lycos.com

unread,
Jun 20, 2001, 1:35:32 AM6/20/01
to loj...@yahoogroups.com
> lojbab lojbab@l...

> Bob LeChevalier, President, The Logical Language Group, Inc.
> 2904 Beau Lane, Fairfax VA 22031-1303 USA 703-
385-0273
> Artificial language Loglan/Lojban:
http://www.lojban.org

mine goes up to 0x260, not 0d600. btw, great book!


Invent Yourself

unread,
Jun 20, 2001, 1:16:27 AM6/20/01
to loj...@yahoogroups.com
On Wed, 20 Jun 2001, Rob Speer wrote:

mi na zgana le do notci be fi xy. .i pe'i lu ganai li'u na'o ralju le


tadji be le nu fanva zoi zoi. if, then .zoi .iepei

Reply all
Reply to author
Forward
0 new messages