Message-Id: <199602071618.LAA03292@locke.ccil.org>
Date:         Wed, 7 Feb 1996 10:46:17 -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: lojban evolution
To:         lojban@cuvmb.cc.columbia.edu
Status: OR
Content-Length: 18409

>I get the distinct impression that
>you are assuming that several dissenters (including me) are extraordinarily
>unsophisticated, and that that lack of unsophistication is the source of
>our dissent.

That is CERTAINLY not my intended implication.  I would go so far as to
say that And, for example is a good deal more "sophisticated" than I
will ever be.  Actually, your typo pins the donkey on the tail.  YOu are
not being UNsophisticated enough, like most of the people who would
learn Lojban, so it may be your lack of UNsophistication that leads to
the problem.  Lojban is NOT intended to be an esoteric language for the
sophisticated, but a model of a natural language, and natural languages
are spoken mostly by the unsophisticated.

>Those participating in the lojban/Loglan schism do not have a unique
>understanding of conflict resolution. Your schism "solution" was, in my
>view, a bloody tragedy; maybe it was the best course, but it is hardly a
>ringing endorsement of your political skills.

Political skills largely have to do with handling the "masses".  I have done
fairly well in that, though I've made my share of mistakes.  But in thie case
of the split, it also involved extremely intense interactions with one person,
a person who claims himself immune to what the masses think, and considers
any attempt to cater to the masses by another as a personal assault.

People with far more talent and experience in conflict resolution have
attempted to negotiate rapprochement between the two efforts, and have
all given up when they got absolutely nowhere with JCB.  He has never
given an inch in negotiations, and when the lawyers tried to do it
formally to keep it out of court, each succeeding effort INCREASED JCB's
demands.

If I blame JCB for the split, it is because a lot of people have tried
to make it otherwise.  Furthermore, I can say that if JCB had not driven
away so many people before I even started, I would not have started the
split, since it was far from my own idea.

I agree, BTW, that it was a bloody tragedy.  I kept up efforts at
rapprochement LONG beyond the time when others had given up.  The result
of JCBs continued inability to even be diplomatic much less seriously
negotiate, made a lot of people so angry that they could not understand
why I kept trying.  There is an enormous history to the defiant
statement "Lojban IS Loglan" approved by the LLG membership a few years
ago - a statement intended in its wording to supplant rather than share
the definition.  You will get further in argument if you start with the
realization that I am the most moderate towards JCB of the founders of
LLG.  I still like and respect the man; others do neither.

>Yet it is management savvy which is needed for the management of change
>in lojban.

A statement that presumes that it is agreed that change is to be
"managed" in the sense most people understand it - which usually
includes fostering as well as controlling.

>Your allegedly "non-political" approach is actually quite political; your
>opposition to "politics" (which you decry) is a defacto quashing of dissent
>so as to rush into publication. You may think you are right, and many may
>agree with you. You are still quashing dissent.

I have never said that I am non-political.  Indeed I said that politics
is inherent in my job - I am the stuckee for the political battles that
are inevitable.  By my doing so, I leave time for others who ARE
anti-political to remain part of an organization that they would leave
if forced to regular participate in political discussion, maneuvering or
decisionmaking.

I will cite my wife as only one example - she was forced to choose
between JCB and her peers/friends in 1984, when he forced matters to an
explicit "vote for me or I am gone".  JCB sent an ultimatum telegram to
her when she had the tie-breaking vote.

She will have NO MORE to do with politics.  At all.  She wants to work
on and use the language, and will do whatever is necessary to keep her
work independent of politics.  You will note her long-continued absence
on this forum.

Yes, I quash dissent.  Are you surprised that I admit it?  We have
people who want to do work and get a language done.  Dissent and
argument prevents the language from getting done.  The LLG organization
is a democracy.  The language design effort is not.  I have assumed near
dictatorial powers, with the agreement, indeed the demand, of the voting
membership, in the interest of getting the job done.  But while I have
dictatorial powers, I have made it a point to seed the destruction of
those powers, so that I cannot become like JCB, so inured to getting his
own way that he thinks that his personal decisions are by consensus.  I
can be overruled on any issue by the membership at LogFest, but they
have made it clear that they want me to ignore the importunings of the
list if it will delay the language (some have urged, rather strongly at
times, that I drop out of Lojban List.  I have been tempted, but if I
did so then there would BE no meaningful dissent.)  I allow consensus to
work on the design, but I do quash dissent by choosing to foster
participation in decisions by those who have shown a will to consent,
rather than to continue battles.

>>The guillotine fell over 3 years ago, but it was not a clean cut, and we have
>>had to try twice more to produce a clean cut.  Hopefully this time it will
>>work, because we are announcing that the guillotineis falling and tying down
>>the neck as much as possible so it cannot move during the fall.  Yecch!
>>what a bloody metaphor! (and a worse pun zo'o).
>
>Interesting choice of metaphor too. Competent guillotining is rather an
>unfuzzy affair, rather a poor choice for analogy to the birth of a conlang,
>I would say.

The births of conlangs have usually been accompanied at some point by
guillotining of dissent.  That dissent then either forms a schism, or
disappears.  If there is no guillotining of dissent, then "change
management" becomes the primary property of the language, and the
language evolves into something totally different, often with a
different name, within a few years.  If there is guillotining, then the
dissenters either form a new schismatic language effort, fade away, or
rejoin and go along with the rest.  Only Esperanto has survived dissent,
and it has experienced all 3 results of its various guillotine
rejections of dissent.

Yes it is an unfuzzy affair.  As a result, the language has an unfuzzy
definition.  If you are on one side of the cut, you are part of the
language community.  If you are on the other side, you are something
else.  If you use the official version of the language, you are speaking
Lojban.  If you deviate too far, then the vast majority of Lojbanists
will reject your usage as non-Lojban.  (Jim Carter can undoubtedly
testify to this.)  And seems to be trying to explore to find out where
that point is %^).

>steven:
>>>Good. I notice there is no gismu for slang:
>>
>>(deleted by lojbab)
>lojbab:
>>Why would we need one.  make a lujvo.
>
>(First you complained about my lack of using emoticons, then you deleted
>both my definition *and* the emoticon, thereby substantially changing the
>meaning of my statement. As 60 minutes proves every Sunday, it is possible
>to make anyone appear an idiot by selectively editing their words. I don't
>need any help being an idiot, I accomplish that task daily even without
>biased editing.)

I either did not see your emoticon or did not read it as having scope
beyond the specifics of the proposal.  Especially when we get within one
day the complaints that there is no word for political activity and no
word for metaphor.  So I took your comment about the lack of a word
seriously.  Why should I do otherwise?

>>I see in such a suggestion a basic lack
>>of understanding of the differences between us and JCB over what constitutes
>>a valid gismu.
>
>I see in your reply either utter cluelessness or feigned misunderstanding.

Well, since I can rule out the latter, that must mean a confessiojn to
the former %^)

>We were discussing slang as an example of post-baselining lojban
>evolution here.  In spite of your planned social disapproval of the
>invasion of the gismu, I predict that invasion of gismu space will be a
>common source of slang in fluent lojban speakers.  Previously, you told
>me that JCB did not approve of gismu slang, when I gave you a reference,
>you then say I fail to understand the difference between JCB's approach
>and yours.  It is *you* who are missing my point.

JCB did not approve of gismu slang.  Slang gismu have made it into the
language only when JCB forgot that it was "not invented here".  For
every slang gismu that made it into 4th edition L1, he ignored dozens of
others, including many that had appeared in print in The Loglanist,
which most people presumed meant official approval since it was an
official publication.

I did not look specifically at the 1989 TLI gismu list, but I know that
in 1986 when I became his dictionary editor, I went through the TLI
gismu list with a very fine comb, and exactly 2 gismu that were not
proposed by JCB were added to the TLI list between 1975 and 1986 (they
were the gismu for "purple" and for "volt" BTW.

>>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.
>
>And Zipf + mimicry = gismu invasion by slang. This is my point. This
>particular issue is an *example* of the kind of problem an academy could
>address. I see that by giving *examples*, I get your attention, (at least
>you are not diffidently dismissing my concerns with baseless assertions, as
>you did previously), but you still just don't get it.

Zipfean effects are not decided by an academy, but by a corpus.  "Slang"
that is introduced intentionally, and which does not exhibit frequency
of usage comparable to other gismu, nor productivity in lujvo making,
will be rejected by the community, not by an academy.  That is not a
baseless assertion, but a design principle.

Indeed, I think that is what YOU are not getting - that Lojban's design
itself includes prescribed means of change.  Violating BOTH the language
prescription AND the prescribed means of change puts a double whammy
against any slang usage of that sort.

Someone asked what happens if people coin grammar that does not follow
the machine grammar prescription.  The answer is that it will not parse
until/unless someone writes a parser that parses it.  NOT necessarily an
official one, but a parser nonetheless.  If that unoffocial parser comes
into general use in order to handle the slang usages it covers that the
LLG parser does not, then de facto that new parser will become the new
language standard.  If no parser is written, then those people who are
into Lojban because of its machine processibility will ignore the
offending slang usage.  Thus only if slang grammar reaches the
credibility that someone wants to process it by machine, will it become
official.  And neither LLG nor an academy needs to decide a thing,
except perhaps at some point of time to decide de jure that a de facto
unofficial grammar will become the official one - at that point a
descriptive rather than a prescriptive process.

>But *you* have a blind spot, which JCB did not have, and that blind spot
>is language evolution.

Believe me - it is NOT a blind spot.  More planning has gone into Lojban
evolution than into the evolution of any other conlang (not saying much
since only Esperanto and Loglan ever had any official policy/plan on
change)

You have undoubtedly heard pc's line "let 1000 flowers bloom" in the
context of many discussions of Lojban.  That is a specific allusion to
change and change policy.  Where we want or expect Lojban to change, we
have provided for Lojban to change without controls.  Where we do not
want change, we have provided a lack of mechanism, which itself will
hopefully prove quite effective at preventing change.  After all - why
promote a change if there is no provision to get the change adopted.  If
the Constitution had no provision for amendment, then the only way to
amend it would be by radical replacement.  We have provided for
amendment in many, but not all, areas.  In other areas we have not
provided for change.  It is quite possible, and maybe even hoped for,
that this means that people will avoid trying to change those areas, and
will use the paths of least resistance to introduce variations and to
solve problems of expressibility.

To put it in terms of your own issue, fuzzy expressions, the deck is
stacked so that you will find it much easier to gain adoption of
proposals that require no new words or cmavo, but merely usage
conventions, lujvo, and fu'ivla.  If that is not sufficient, then
coining experimental cmavo that fit existing selma'o will provide the
second level of pressure valve, since all you need is a new table and
not a new machine grammar/parser, in order to process that new usage.
Only if you cannot achieve your expressive goal with those means are you
likely to resort to a experimental usage that demands grammar change (as
I gather And's proposal would), because someone will have to write a
parser or the machine-oriented people will ignore it (which would be
particularly devastating to a proposal in fuzzy logic expression, I
should think).  LLG will not be writing a new parser for a long time -
that puts tremendous pressure to find solutions in the existing
mechanisms of the language, because language use, by corollary to Zipf,
always takes the easy way out.  Presto, language evolution of fuzzy
logic expression without any academy.

>Put aside your preconceived notion of the flaws in JCB's text, and just
>reread it for the *ideas*.

I assure you that I did so.  There were few new ideas in 4th edition,
other than the new Chapter 7, and most of them were poorly thought out.
His language change philosophy in Chapter 6 was not new, and was
explicitly rejected by the founders of LLG as being unacceptable to the
community.  I think we came up with a better idea.

>That's why I posted the JCB excerpt; it addresses precisely the issue
>you are glossing over.  JCB is right on about change, conflict between
>different types of lo??an-o-philes regarding change, and language
>development in general.

I know how JCB understands the words he wrote.  You apparently do not.

JCB has NO provision for slang.  He has an Academy, and has said in
print that he will publish no Loglan that violates what he considers
official Loglan.  He is change tolerant only in that if you want a
change, you can file it in triplicate with the Academy according to very
strict guidelines, and the Academy will deliberate.  No promise that it
will decide.

Jim Carter certainly understands what JCB meant, as does Bob Chassell,
who was Lognet editor when Carter came up with what is now labelled
Nalgol, but which was no more radical than your proposed slang
procedures.  Carter wrote lots of Loglan text using his "slang", far
more than had been written by all other Loglanists put together, came up
with formal language descriptions that were at least as carefully
writeen as TLIs, wrote his own primer, and submitted materials to Lognet
for publication regarding his suggested changes and usages.

Chassell published a few of them, and JCB fired him as editor even
though it was explicitly in the TLI wirtten and approved policies, and a
contract that JCB had signed, that the Lognet editor was not subject to
JCB's direction.  That was the fight the specificallly precipitated the
1983-4 war (thought there were some precursor battles).

As part of the war, JCB explicitly and personally convened an Academy
composed of Jeff Prothero, pc, and himself, and went through all of
Carter's proposals that they recognized (he had made a list), rejecting
most of them.  There was no public debate, there was no process that
resembles slang absorption into the mainstream language, and no
provision that the changes could percolate along unofficially for later
ruling.

THAT is what JCB's pretty words in Loglan 1 mean:  There can be language
change, but you propose, and he disposes, but only if he chooses to take
official notice of your proposal.  There is NO unofficial TLI Loglan.
It is either official, or it is in violation of his now-thrown-out
trademark (Where did you think the trademark issue came from?  Lojban
started out as nothing more than "slang Loglan" of exactly the sort that
you are proposing.  We reinvented the words and the grammar successively
because JCB specifically said that he would sue us if we published
standard Loglan without his vetting, or if we used any deviations from
standard Loglan at all.)

Now in practice, he has undoubtedly allowed some slang to appear in his
publications, but this is "not careful editing", or maybe even
"subversive editing", and NOT JCBs or TLIs policy.

JCB's practice allows for some grammatical evolution, in theory.  But
his concept is more constrained than LLG's.  Those words you quoted
fairly explicitly rule out slang grammar, minimize slang cmavo, and
tolerate or encourage slang brivla.  LLG agrees with all of these,
noting only that what he says about coining of content words usually
falls within the context of using existing roots.  New roots are rarely
added to a language except by borrowing (xerox and kleenix being a
couple of recent exceptions in the general language) or by abbreviation
(laser being an example).  Lojban has procedures for all of these.

>You JCB-detractors are <peha> throwing out the baby with the bathwater.

And one JCB apologist needs to learn some JCB history, so he can join
debate with another, quite practiced, JCB apologist.  (I think And can
certainly recall several times when I put down one of his proposals with
lu'e"this is the way the language is, because that is what JCB said, and
I agree with it".)

I don't need to disagree with one word that JCB wrote in Chapter 6. I
have called Lojban "Loglan" for 9 years now because I largely accept of
JCB's philosophy.  Loglan is the language that is defined by JCB's
words.  We practice what JCB wrote far better than he does.  But TLI
does not follow JCB's written words.  Lojban IS Loglan.

lojbab