Message-Id: <199602061213.OAA08964@xiron.pc.helsinki.fi>
Date:         Tue, 6 Feb 1996 07:12:11 -0500
Reply-To: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      Re: <polti>
To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 14196
Lines: 283

>The absence of a gismu for politics is merely
>one of many holes that are in gismu-space.

I am sure that if we agree on a definition, that I can come up with a lujvo for
 it %^)

>But JCB discusses the construction of slang gismu
>independent of the formally constructed list in chapter 6 of loglan 1 4th
>ed.; I was curious regarding lojban central's policy regarding slang gismu,
>and I am still not sure what it is.

The policy is against slang gismu, because the quality of research done is
usually abysmal (not thatthis is specifically true for you), the people
usually have not tried to address the issue with lujvo, the concept used
is very culturally specific, and most of all, there is little tendency to
mark the slang as such, so that unlike xVV/cVVV and fu'ivla space, someone
who sees a slang gismu does not know it is slang, and hence will try to
remeber his  gismu list.  memorizing the gismu list has been an icon of
Loglan mastery since 1976 when JCB set forth his flash cards and said that
97% mastery was the standard.  A fluid gismu list cannot be mastered to
the 97% level, and indistingusiahble slang renderes the list fluid.

I have devised aconcept for a word-space solution to the slang-gismu
question that Cowan is reviewing privately to see if we even want to
consider such a thing atthis late date.  If it works, then it will be
discussed.

As an example of why we don't like slang gismu, i suggest you look in
L1  mod 4 for one such that JCB incorporated into the language officially.
"nukle" is in Appendix B, but he uses "nukli" in other places (footnote 1.21
for example).  Appendix B has gismu "tulpi"; Appendix E fu'ivla "tulpa".
I will not that for JCB, a slang gismu that has not had the 6 languages work
done is a fu'ivla, and he does not object to fu'ivla treading on gismu space.

But he forgets that one purpose of GMR was to unpack the heavy concentration
of gismu in some parts of the alphabet that made them hard to learn and
hard to distingush in speech.  Lojban has a policy that prevents some kinds of
gismu list packing, but the nature of slang is that if we permit slang gismu,
then some will descriptively enter the mainstream language, even though they
might violate the design policy that surely most Lojbanists will not memorize.
Things that most Lojbanists will not know how to do correctly are not fertile
areas to allow slang to develop.  That is why we have free slang use of
Type III fu'ivla, but strongly discourage the error prone coing of Type IV
fu'ivla -a t least till there is some sense that enough words have been
made by such a pattern that there is a chance that people will adopt
good patterns by mimickry since doing so by analysis will not be well-taught
or well-understood.

I am posting the old review of L1, BTW since addresses the changing language
issue even back then in 1989 when it was written.  This is not a new policy/
philosophy of LLG, but one that dates to our founding, and it USED to be well
understood and accepted when most of the community had memories of what we
were trying to do in splitting from JCB.

>If I were to take this slang and want to make it part of the official
>lexicon of lojban, then *that proposal* ought to be in lojban. Slang is
>spontaneous and ought be unregulated, in my view.

Slang is by definition unregulated and unregulatable.  But if someone
creates a slang brivla in cmavo space, assigns a slang rafsi to a fu'ivla,
and renders avrious other things, it is a foundational deviation from the
language - the sort of thing that would NOT be done spontaneously in a
natural language, at least not often enough to be noticeable.

I agree that slang should be spontaneous.  But it should be spontaneous among
speakers of the language.  We do not consider a botching of English by a
non-fluent speaker to be "slang" whether or not it was an intentional error.
It is simply an "error".  It is not "slang" unless it occurs spontaneously
among the native speakers.

Lojban has no "native speakers", but those who talk on the list
predominantly in English are not their counterpart.  Thus a proposal made
by an English speaker in English for use in Lojban is an outside attempt to
"inject" slang into the language.  It has a reasonable chance of working, too,
unlike natlang counterparts.  And I do not respect such attempts.  If
Goran introduces his gismu for Croatia in his Lojban, I have no objectuon
(though I will not support the slang and will try to convince him to use
a different formulation) - it is at least being done the right way.

>Are you saying that slang usage is verboten now, but will be tolerated
>after baselining? I don't follow what you are saying, as it appears to
>contradict your expressed tolerance of slang elsewhere in your postings.

Slang usage is tolerated now mostly as a means to talk about proposals.
And usually it occursin English language examples of such proposals.
It is thus a necessary evil to get the discussion business done when we
need to make decisions quickly.  Thus time pressure itself is generating
an unnatural amount of slang in unnatural places for slang to crop up
spontaneously (your quote from JCB notes that cmavo slang is rare and
new
grammatical forms in slang even rarer, yet this is where MOST Lojban slang
is appearing these days).

If slang is appearing in Lojban text, I would have little way of knowing.
Presumably from his comments, Jorge's proposals were derived from actual
slang usage on his part, though it may have been originated non-spontaneously
in response to his review of the refgrammar. Goran's coining was apparently
spontaneous.  Almost no other "slang" other than fu'ivla and nonce lujvo
are being coined these days, to my knowledge.

So my non-tolerance of slang is to the introduction of non-spontaneous slang
deriving from prescrriptive discussion (especially such discussion in
English).  Any word that is first mentioned in English discussion is defacto
non-spontaneous LOJBAN.  Any word that first comes up even in Lojban during
metalinguistic discussion is also probably not "slang" because its
spontaneity is not of the nature normally associated with slang - which is to
use first and define later if necessary.

And I think we have too few people who are skilled in Lojban discussion to
fairly evaluate slang before deciding to adopt it or not adopt it at this
point.  Presumably after the baseline, (one would hope) there will be a
 significant increase here on the list to conversation in Lojban rather than in
 English,
and theskill level and breadth of skill level will increase rapidly.
If that happens then I will be more trusting of slang that "arises", and
not see a subverting conspiracy to subvert the baseline in every coining %^)

But I still hope that Lojban slang arises slowly, because the rapidity with
which slang arises is a measure of the quality of the design.  Too much
slang means that the baseline design basically WAS incomplete for all the
extra time we spent on it.  This is different from saying that the language
is incomplete, a truism until the content words expand to fill semantic space,
hopefully predominantly as lujvo, since that was the way we stacked the
design.

>I purposely chose <polti> as a slang usage, because it mimics gismu, and is
>decidedly not a part of the formal language. Perhaps you I think speakers
>are likely to readily generate slang gismu. Surely you do not propose to
>squelch this spontaneous activity *after* baselining!! Perhaps you failed
>to understand what I was about. Here is where I am coming from:

There are some things that people do not do when they generate slang
spontaneously, as JCB notes.  they don;t generate new grammatical constructs,
they seldom add new cmavo.  I could also add that they seldom add words that
violate the phonemic patterns of the language, as they pertain to morphology.
Thus a word ending with -ly in English will usually be an adverb, even in
slang.  Jabberwocky was reasonable evidence of this.

I would like it to be that coining a new gismu adhoc in the standard pattern
is just such a "thing that is not done", enough to try to find ways to
make such a thing unnecessary.  Until recently social pressure would have
been enough, but it is clear that the social mileau of the Lojban community
is no longer totally dominated by the anti-gismu people - the ones who would
vote "no" on your "polti" even if the lsit wer not baselined because they
believe that there are already too many gismu.

>1. Slang is a vital omnipresent feature of living languages.

Agreed.

>2. Slang serves as an important source of valuable new ideas for accretion
>to the formal structure of a language.

I don't see evidence of this, and it seems to contradict what JCB wrote as well.
Formal structures don;t change from slang, though they might from mimicry of
foreign language patterns when the foreign language is not stigmatized
(hence Frenchisms in English like "deja vu"). Language evolution in a broad
 scale may be the result of interactions between language communities.

>3. Much slang is mimicry of legitimate constructs in the language.

True.  But what constitutes "mimicry" is following a pattern in the words
that is accpeted for slang, and not following a pattern that is not accepted.

But this is a weak negative, and the main reason I want to start out with
a social stigma against new gismu.  At some point it won't matter because
the language vocabulary will be so large that the gismu are a small minority of
the lexicon, and most people will feel little need to coin words at all.
Atthat point, I can take my chances with slang gismu, knowing that they will be
rare (because the gismu pattern will not be so noticeable in the language)
and there will be less pressure to make them because theconcept will already
be represented by a word in the lexicon.  In the latter case, even a novice
coining a gismu will not sway the masses of fluent speakers to drop the lujvo,
UNLESS Zipf's Law makes the shortened form justified by frequency of usage.

>Apparently my intent was not clear to you. Still, there is a hole in lojban
>where <polti>, (or the 7-language weighted algorithmically derived
>equivalent) ought to be.
So make a lujvo.  UNtil you have made many lujvo to cover the various meanings
of "politics", I \will be uncertain even what meaning you assign to polti.
I recall trying to analyze the English "tyranny", and I think we came upo with
at least 6 or 7 lujvo in such discussion.  Politics and sociology are fields
where words are heavily polysemous, and should be defined as lujvo.  I
consider a gismu for  "politics" to be as risky as a gismu for "fuzzy" where
it would attempt to compass the full range from my daughters fuzzy bear to
fuzzy logic.  It is to me likley to be an English way to divide up semantic
space and not a Lojbanic way.

(Somewhere I have a posting from Don Harlow on the Eastern European concept
of "political activity" which is alien from both the American one and from
JCB's.  I do NOT want to see one Lojban word attempting tocommingle those
different concepts except perhaps as a "class/category" word which can clearly
be a lujvo.

>There is a tone of "look what those horrible speakers are doing to my
>beautiful language" about your disapproval.

No - it is more a "look what those horrible NON-speakers are doing to the
language".

>Will your objection to such
>subversive language terrorism (that is, slang) continue *after* the
>baseline?

If it is terrorism - i.e. done by intent to force change rather than simple
fulfillment of language needs, then yes.  But otherwise my opposition
will decvline proportionate with the growth of true language competence in the
community.

>Will we have a language
>police?

Hardly.

> Or a "Miss Syntax" column in the lojban journals to advise us on
>proper language use?

I wouldn;t doubt it.  But I hope that if written for speaking Lojbanists that
it will be written in Lojban.

>The point here is I am deliberately introducing a (very small) problem of
>the sort that I believe will inevitably occur after baselining to clarify
>for you what I see as being an important issue.

And we have too many present problems to try ot solve post-baseline problems.

>How will this be handled?

Ad hoc.  That is how natlangs handle it.

>Tolerance and anthropologic note taking?

For the scientists among us, yes.

> Consideration by the lojban
>academy, when this illustrious body assembles in 5 or 10 years?

I can;t speak for them.

>Pistols at
>twenty paces at dawn?

Not too easy on the net.

>Shunning of the offending coiner of slang?

In my case probably, since I have found slang usages particularly hard to
cope with because I can;t distinguish them from errors.

>horror just a pre-baselining stance?

When it is clear that slang is the result of true need among the language
competent and not a "movement" (nonspntaneous) and not an "error", my stance
will soften.

>No, you are both wrong. Incompetently managed change can be disasterous;
>competently managed change can be productive. Management requires skills
>which that only few people possess, but which many people either think they
>have or think are trivial.

If the skills are rare and the leadershiop of LLG does not feel it possesses
the skills, then we are wise to avoid trying to manage change as opposed to
squelching it.

>Many people disparage management, politics, and
>bureaucracy both because they have had a good dose of incompetent instances
>of these,

Abd these are the people who we want to convince to learn the langauge,
 therefore we better not subjectthem to more of the same.

>>I'll keep listening, and hope you did not mean what I understood.
>
>Good. I notice there is no gismu for slang:

Why would we need one.  make a lujvo.  I see in such a suggestion a basic lack
of understanding of the differences between us and JCB over what constitutes
a valid gismu.  In Lojban that is a "root" - something useful in compounds or
which cannot be represented in compounds.  For JCB it was an attempt to find
semantic primitives i.e. basic  ideas., with a heavy load of Zipfean analysis of
 other language built in.

lojbab