Purpose of the gismu list

[Logical Language Group <lojbab@access.digex.net>]

Missed sending 1 message yesterday.  I'm sure I'll be forgiven %^)

>I think I misinterpreted the purpose of the gismu list, then.  Is it
>supposed to be a guide for how you hope the language will actually be
>used?  I thought it was the sort of thing that someone will go back in
>20 years or so and revise based on what people actually use in practice;
>but that's not entirely true if there are places that people will never
>bother to fill.
>
>(On the other hand, there might be places used only in descriptions.
>For instance,

The purpose of the gismu list and all of our other documents is to be as
thorough and lasting a language prescription as possible given time and
circumstances.

The more thorough the prescription, the more likely that a large
category of possible Lojbanists will come to try to learn and use it.
Underspecification may cause/allow some people to guess, but for other
people, they just avoid trying.

We know there are minor flaws (and perhaps some major ones?) in the
language, and more specifically in the details.  As the language passes
from its prescriptive to descriptive phase, these will be discovered,
and usage will then supersede the prescription if the problem causes
sufficient discomfort to a language user that he ignores some rule.

In addition, fluent speakers may tend to make consistent errors based on
what seems natural for them in the language.  Consistent errors will
come to be understood as the norm, and in the descriptive phase, will
become the new dictionary entry.  Usage by fluent speakers and writers
will always have more influence on me than even a carefully argued
proposal for change, unless it deals with a critical design element for
the language like the logical aspects, where we have reason to believe
that natural inclinations might not (at least initially) be valid for
the purpose of the language.

In some cases, the gismu places serve as a fairly detailed definition of
the word.  Thus in your example:

>jakne x1 is a rocket [vehicle] propelled by jet expelling x2 carrying
>payload x3

jakne is clearly defined as a vehicle with a payload and a jet.  If it
is missing one of those things, it is probably not a jakne, even if it
might be described as a "rocket" in English, and you may want to look
for another word.  This helps you separate the meanings of the words
from their English keywords.  My experience is that the place structures
have been the most effective way to show such distinctions in meaning.

Since a rocket always has a payload and a jet, even if they are not
specified, you may then make inferences about the context in which the
word appears.

One of the best examples I think is the word xamgu
x1 is good for x2 by standard x3.

It has even changed my English - I am now quite prone when someone uses
the word "good" to instinctively ask "for whom", and to recognize that
the standard of goodness is always relative.  Yet I doubt that I have
ever used or specified the x3 place of xamgu, and I rarely do so with x2
either.  Nor do I in English - BUT I think about what the word means a
bit more and thus am more careful in my usage.

For a while we went overboard, as I mentioned recently, adding a lot of
places that made people very certain about what the word meant, and to
think careful how they applied the word, but really added nothing useful
- the discussion of klupe/screw shows the boundary at which we came to
feel usefulness turns into pedantry.  Leaving the place in that word,
allows someone to identify a "klupe" in something that may not
necessarily be called a "screw" in English - e.g. a "bolt" because we
have specifically mentioned threads.  Likewise, someone may recognize a
mruli in a rock wielded by someone's arm.  But you no longer have to
define the blade (and shape) and handle of a shovel because other than
being bladed in some sense, the focus is shifted primarily to what it is
intended to dig.

Yet, I doubt that most people will need to specify the parts of a mruli
or a klupe, nor what the canpa is supposed to dig.

Will these places disappear with time?  I'm not sure.  I doubt that they
would disappear before we could write a Lojban-only dictionary, even if
they weren't used.  More likely is that some place structures will
change, with higher numbered places that are more useful moving to the
front.  And such moving will make it easier for us to focus on the
always-omitted place, perhaps allowing us to look for usage that
indicates that it is or is not playing a real role in determining the
actual meaning in usage of the gismu.

But the latter is how I see meanings changing - not by fiat from
dictionary writers or by proposer who think they see a better way.  We
will "let usage decide".

lojbab
----
lojbab                                                lojbab@access.digex.net
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273
 For the artificial language Loglan/Lojban, see ftp.cs.yale.edu  /pub/lojban
    or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/";