Date:  Fri, 5 Dec 1997 10:59:18 -0500 (EST)
Message-Id: <199712051559.KAA20873@locke.ccil.org>
Reply-To: And Rosta <a.rosta@UCLAN.AC.UK>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: And Rosta <a.rosta@UCLAN.AC.UK>
Organization: University of Central Lancashire
Subject:      On logji lojbo discussions
X-To:         LOJBAN@cuvmb.cc.columbia.edu
To: John Cowan <cowan@LOCKE.CCIL.ORG>
X-Mozilla-Status: 0001
Content-Length: 10453
X-From-Space-Date: Fri Dec  5 10:59:47 1997
X-From-Space-Address: LOJBAN@CUVMB.CC.COLUMBIA.EDU

[I can't tell from message headers whether messages come from
Lojban list or not.]

Lojbab:
> >> >and John has proposed that
> >> >a nu is an abstract thing, while usage indicates that a nu is
> >> >a concrete thing
> >>
> >> And Lojbab thinks it is ambivalent because it depends too much on what you
> >> mean by abstracty and concrete.
> >
> >I can't make any sense of that statement.
> >It seems that you don't have an articulable view of what "nu" means.
>
> Actually, I think I have articulkated it, but I have never managed to
> understand what is unclear to you.

There are several mutually incompatible possibilities for what nu
could mean, and each one of them has been articulated by you at
some point in this recent debate. I will try to post a summary
of the options, at some point soonish.

> You seem to have this problem only with
> "nu" and not with the subevent types, BTW ( or at least you do not mention
> them), which adds to my confusion of the nature of the issue.

My policy is first things first. Once I've understood what the
lojbo bigwigs think nu means, I can then see whether it seems
to make sense in relation to other things (such as the subevent NU).

> >> It also cannot be said that when the possibility of confusion is
> >> not even being tested on the majority of people skilled enough in
> >> the language and in logic to become unconfused.
> >
> >???? I can't parse that!
>
> Lost my train of thought.  Most of the people who are skilled enough in
> the language to be able to address the issue aren't even reading the list,
> much less this discussion.  Henceit cannto besaid that this is an issue for
> "most" or that "most" will understand the outcome.  "Most" will never see or
> hear of the discussion.

I forget what context this point was being made in. At any rate, I
do assume that anyone who has the time and desire to participate
in these discussions could. (There may be the odd exception of
someone like Goran who possibly can't get net access.)

I would like the fruits of the discussions to be more widely
disseminated, and I hope that we are on the brink of resolving
to try to achieve this. But so long as LLG is unsympathetic to
them, all the participants can do is plough on among themselves.

> >> Furthermore, we have taken the lexicographers view that all words
> >> are to some degree polysemous. The meanings of a word in its
> >> various usages could be mapped to semantic space as a kind of
> >> scatter-plot rather than as a strict point, and the word "meaning"
> >> is an area of that semantic space that encompasses all the usages.
> >> (Of course this definition implies that usage determiones meaning,
> >> which is why I insist on having more usage before we decide what
> >> people mean.)
> >
> >If this is what you mean by polysemy, then let us speak of
> >homonymy instead.
>
> No that is not what I mean by polysemy, but it is a form of polysemy.
> And it is plausble to me that some of your claimed examples of
 homonymy/polysemy
> are just rather more scattered usages.

It is a possibility in some cases, and one that I have recognized
all along.

> >Suppose {xlura} were a homonym, xlura1 meaning "flower, bloom",
> >and xlura2 meaning "flowering plant". Then {lo`i xlura} would
> >be ambiguous, according to whether it means the set of all
> >flowers, or the set of all flowering plants.
>
> But as a tanru modifier (seltanru) I suspect that xlura DOES encompass
> both meanings as well as a large number of other plausible ones having
> mildly to do with flowers.  xlura does have different meanings depending on
> its gramamtical role.  You seem to be concerned about polysemy of the bare
> gismu as a selbri, which is a fairly limited question.

Tanru modifiers do not affect truth-conditional meaning. To find
out the sense of a gismu, one needs to examine it as a selbri.

> >> compounded by sumti raising errors that us sloppy English speakers
 habituall
> >> promulgate.
> >
> >You keep on mentioning these alleged sumti-raising errors, but
> >if at any point you showed them to be relevant, I missed that
> >posting. Could you reexplain? I can't see what they have to do
> >with ni and jei.
>
> Here are excerpts from several messages in the last couple of days on this.
> first last
>
> Jorge:
> >{jei} is defined in the refgram both as a truth value, and as an
> >indirect question involving truth values.
>
> Lojbab:
> >I claim that the refgrammar does not define the second.  I will agree that
> >Chap 11, 6.3) appears to contain an unmarked sumti-raising which would
 require
> >a tu'a on the x2 to be ideal.  I think that we will find that there are
> >other usages in the refgrammar that have hidden sumti-raising which makes
 them
> >less than perfect examples of the language in light of the given English
> >translation.  But the fact that an example is not entirely consistent with
> >the text describing the principle does not make the text definition
 incorrect.
> >
> >I can accept that the refgrammar examples will have logical flaws, especially
> >with regard to their given English translations, while upholding the
> >refgrammar as a baseline standard.
> >
> >(Resolving an apparent inconsistencey between the rst of the section and
> >6.3) merely requires that I analyze that the English transaltion given for
> >6.3) does indeed reflect a sumtiraising of an indirect question.  Neither
> >the translation nor the text discussion indicate that the indirect question
> >was supposed to be the meaning of the jei.

I now understand what you meant by sumti-raising.

I don't recall the message you quote from, but it seems an
entirely satisfactory resolution to the ni/jei question (assuming
that ni is resolved in the same way as jei).

[big snip]
> And of course, if it had indeed been put that the refgrammar had usage
> errors due to sumti raising that were occluding the definition of jei
> rather than that there were two contradictory definitions of jei in the
> refgrammar, I think I would have understood much quicker.  And since one
> of the examples in question involved the x2 of djuno, where many
> previous debates have led to a ledu'u being the non-sumti-raised value,
> that this was an unmarked sumti rasing should never have been in doubt.

What Jorge and others pointed out was an apparent inconsistency
in the refgrammar. That this inconsistency was due to sumti-raising
errors was only one possible way of resolving the problem. In
the way I read the debate, I read you as denying that there was
any inconsistency - i.e. as claiming that all relevant bits of
the refgram are pukka. Therefore, as Chris suggested, it seemed
that the only option was that ni/jei were homonymous. Only then
did the homonymy solution get propounded.

Also, calling the erroneous usage of ni/jei "sumti-raising" is
unhelpful, since using "tu`a le ni/jei" still does not fully
capture the intended meaning. Others called it "indirect question"
which is far clearer.

> >> I think that usage will gravitate towards a specific meaning,
> >> which meaning for jei is the one where Ch 11, 7.3) wil come to be
 considered
> >> invalid sumti raising.
> >
> >Which one is that? (Our web connection is down again so I can't
> >check.)
>
> mi djuno lejei la frank cu bebna
>
> It should have been
>
> mi djuno tu'a lejei la frank cu bebna
>
> or
> mi djuno fi lejei la frank cu bebna
>
> to not be sumti rasing since the x2 of djuno is normally a ledu'u abstract
> sumti

Right, but to actually get across the intended meaning, it should
have been

  mi djuno lo du`u xu kau la frank cu bebna

> The refgrammar in discussion only mentions one meaning.  The other
> "meaning" is inferred from the examples.  As to whether jei has
> always had the "whether" meaning, I could ghardly sa, without looking
> for the examples in usage (of which there have been few I suspect since
> it isn't the most useful thing in the language - and most usages will
> date to before tu'a entered the language in any event.

Had jei meant "whether" it would have been VERY useful. If
usage of jei has been limited it is because (a) its true
"truthvalue" meaning is pretty useless, and (b) most users
realized that using "jei" to mean "whether" would be erroneous
(even though I am sure that "jei" was created with the intention
that it mean "whether", but then given an incorrect but
subsequently binding definition).

> >Note that that ni/jei thread was asking which of the two meanings
> >of ni/jei was correct. IIRC noone was arguing that the homonymy
> >answer was desirable; it just seemed to be the de facto situation.
>
> And it may be that the whole debate would have been preempted if John hads
> been reading it or if I had picked up on what the issue really was a lot
> sooner.

Doubtless.

> >John is of course an authority on the baseline, but Jorge is an
> >authority on unsolved problems (= valid questions to which there is
> >no established answer). As the baseline is well documented, thanks to
> >John, and the unsolved problems are not, I think the knowledge stored
> >in Jorge's mind is more precious than that stored in anyone else's.
>
> Maybe we should ask Jorge to document these clearly, along with his
> thinking as to what multiple options have been presented in usage or
> refgrammar. Then the various people can look over the list and see if there
> is agreement that they are unresolved before we go trying to resolve them %^)

I am wholeheartedly in favour of this.

Note that I do not then go on to advocate any change of the
baseline, or anything like that. Rather, I would like us to
discuss unresolved issues, and what the best resolution would
be, even though the results of the discussion are non-binding
(at least not until/unless usage entrenches them).

Are you reassured by that? Are we moving towards agreement?


> If people get so carried away with the logical structure that they
> don't choose the right lexical words, choices, they will not be
> understood.  I am not saying this will happen, but Iam doubtful
> that analysis will prove as effective in finding problems as
> usage is.  I base this on the last several years, where all that
> malglico pidgin fumbling has uncovered errors and resolved them
> more quickly and easily than the analytic approach has done.

Problems tend to be discovered by alert users, and resolved by
analytical discussion. Jorge is so good at finding problems because
he is such an alert user.

--And