Re: <mahi> and definitions

["Robert J. Chassell" <bob@GNU.AI.MIT.EDU>]

   >A place is a part of a gismu.  When you speak, you yearn to fill it.
   >It is in your mind and in the mind of your listener, even if it is
   >filled by {zo'e} or is not said.  (Or at least, this is what should
   >happen; I have a hard time remembering the first place of a gismu, let
   >along the fifth.)

sbelknap@uicvm.uic.edu:
   But if you can't remember where and if a gismu has a standard place, how
   does this bring the notion of a standard closer to the attention of speaker
   & listener? Its like trying to remember if <le stylo> and <la chat> in
   French are masculine or feminine.

The presumption has been that fluent speakers of Lojban will have
learned the language!  :)

   Why should "standard" be more important than the other modifier concepts
   embodied in the selmaho BAI?

I am not claiming that it is or should be; only that concept in
selma'o BAI is likely to be further from the attention of a speaker or
listener than a concept that is in a place.

   >As for the current set of words with `by standard' --- I am not too
   >inclined to make many changes.  For one, the current inconsistencies
   >will enable us to find out in five or ten years if my Whorfian
   >prediction really is true, that fluent Lojban speakers and listeners
   >will more often consider a standard when they speak or hear a gismu
   >that has a `by standard' place, than when they speak or hear a gismu
   >that lacks such a place.

   This is an interesting point which I hadn't considered.  Perhaps if
   the etymology of the places of the gismu were explained, things
   would make more sense.

I fear it would take too long for Lojbab to write explanations of them
all, even if he could remember accurately!  Here is an example.
Editing this after writing the next couple of paragraphs, I can tell
you, it takes some time!

{xamsi}, ocean or atmosphere, was mentioned recently, on account it
has an `on planet' place.  I think its definition is excellent.  Here
is my understanding of the reasoning:

Firstly, {lalxu} handles liquid bodies within a landmass.  {lo lalxu}
can be of any size.  Secondly, nature calls for a word for the concept
of a layer that covers all or much of a planet.  However, such a
concept can apply not only to a water-based, liquid ocean, but also to
a magma ocean or a gaseous methane ocean.  The covering fluid need not
be liquid.  Hence the inclusion of places for the composition of the
fluid and the planet to which reference is made.  We need them.

* lake,
    x1 is a lake/lagoon/pool at site/within land mass x2
    /:/
    /=/ lalxu (la'u)

* atmosphere (sea, large partially or fully enclosing layer of a planet),
    x1 is a sea/ocean/gulf/[atmosphere] of planet x2, of fluid x3;
    (adjective:) x1 is marine
    /:/ [also atmosphere (= varxamsi, varsenta)]
    /=/ xamsi (xas)

Of course, I have a quibble with the definition for {xamsi}; it should
include `star' as well as planet, so one could talk about the
chromosphere of the sun as well as the ammonia layer of Jupiter.
Indeed, when people start talking about stellar morphology, I expect
the definition to extend that way.  Furthermore,  cool spots on a
solar surface (`sunspots') may end up being called {lalxu}, as lakes
of coolness within a hotter plasma.

All in all, I am very impressed with the definitions of gismu.  Mostly
they are much better than what I would write, were I of the
temperament and circumstance to write a dictionary.

   But wouldn't it be a purer test of Sapir-Whorf to "standardize the
   standards" by either forcing the use of ma'i for all gismu or
   including a standard place for *all* (without exception), and then
   seeing what patterns of usage emerge?

Lojban *already* includes the use of {ma'i} for all gismu.  All of
selma'o BAI are there for all gismu.  (Just in some cases, {ma'i}
would be a reduplication.)

If we had enough fluent Lojban speakers, we could conduct this test.
My point remains the same, that people are more likely to use notions
that are built into a gismu than to use notions from selma'o BAI.

Thus, a modern automobile designer is likely to have `emergency brake'
as part of his or her concept of `automobile'; but that designer may
not have `easy-to-handle knobs' as part of the concept.
`Easy-to-handle knobs' are like selma'o BAI, a part of the definition
a designer can readily forget; `emergency brake' is like a built-in
place.  You got to have it, unless you go to considerable effort (as
with using {zi'o}) to do without.

   It would be interesting, for
   example, if native Deutch speakers used standards when talking
   about <blanu> and native Italian speakers used standards when
   talking about <cinmo>.

You can set up a research protocol to do this with the current Lojban,
either by retraining the Lojban speakers to use a redefinition of
{blanu} and {crino} that includes a `by standard' place, or by asking
that they consistently use {ma'i} in the experimental setting.

    Robert J. Chassell               bob@gnu.ai.mit.edu
    25 Rattlesnake Mountain Road     bob@rattlesnake.com
    Stockbridge, MA 01262-0693 USA   (413) 298-4725