predicates of motion - late response

["Dylan P. Thurston" <dpt@abel.MATH.HARVARD.EDU>]

ki'e. lojbab le di'e danfu
mi cusku di'e
 > >Or take the motion verbs expressing a manner of movement.  I count the
 > >following:
 > >
 > >cadzu   x1 walks/strides/paces on surface x2 using limbs x3
 > >bajra   x1 runs on surface x2 using limbs x3 with gait x4
 > >stapa   x1 steps/treads on/in surface x2 using limbs x3
 > >vofli   x1 flies [in air/atmosphere] using lifting/propulsion means x2
 > >cpare   x1 climbs/clambers/creeps/crawls on surface x2 in direction x3 using > >                        x4 [limbs/tools]
 > >farlu   x1 falls/drops to x2 from x3 in gravity well/frame of reference x4
 > >sfubu   x1 dives/swoops [manner of controlled falling] to x2 from x3
 > >plipe   x1 (agent/object) leaps/jumps/springs/bounds to x2 from x3 reaching
 > >                height x4 propelled by x5
 > >
 > >IMHO, these are similar predicates and should have similar place
 > >structure.  But:  only some include the medium; {cpare}, uniquely among
 > >all motion verbs, includes a direction; {bajra} but not {cadzu} includes
 > >a gait; and {farlu}, {sfubu}, and {plipe} but not the rest include
 > >source & destination.  As a result, it's very difficult to talk about
 > >someone falling down an infinite pit (consider, for instance, Alice
 > >falling down the rabbit-hole in "Alice in Wonderland"[1]).

la lojbab cusku di'e
 > At one point they all (?) were identical to klama in place structure.
 > But the lean gismu people wanted redundancy eliminated.  So now you
 > bajra klama or cadzu klama.  Indeed, I think there are remnants of the
 > old place structure in some examples in the draft textbook.
 >
 > In other cases, places were added to account for cultural or
 > metaphorical uses of words.  You can walk on your hands - so we added
 > thhe specific limbs.

Yes, absolutely.  I didn't mean to argue for identical place structures,
just some justification for the differences.  The limbs I can see being
very handy.  (See below...)

 > The gaits of 4-legged animal motions apply to
 > various degrees of running (you could say that walking is a specific
 > kind of running gait in a 4-legged animal).

Questions here:  {xu le nu cadzu be fi remei cu nu bajra mu'i ma} (Is
walking a kind of running gait for 2-legged animals, too, and why?)

What's an example of a gait (say, "gallop"), other than by le'avla?  Why
can't the different ways of running be handled with tanru or lujvo off
{bajra}?

 > Direction was added to
 > climb when we expanded it to include clamber and crawl, which tend to be
 > related etymologically in many languages.

I'm not sure I entirely understand your point.  There's already a
surface:  wouldn't {cpare le loldi} naturally mean "crawl", while {cpare
le sraji} ("crawl on the vertical (surface)") would be "climb"?  (You
could be moving horizontally, but I think those rare cases would also be
covered by the English "climb", though perhaps "traverse" is more
appropriate.  I would recommend a spatial tense, but none seem to be
appropriate.  How would you say "climb across the rock face", anyway?  I
didn't see any appropriate spatial tense.  Maybe {pinta cpare le rokci
sraji}?)

Is "clamber" just {juxre cpare}?

 > plipe was always a pain
 > because its English and other language equivalents tend to be used for
 > both jumping/springing up jumping over, and jumping from/to. the
 > specificity of the place structure was the best way we could think of to
 > clarify the core meaning and exclude those things we did not think fit
 > the core concept (or force them to be lujvo).

I'm still a little uncertain here.  Which meanings did you want to
exclude with {plipe}?

In any case, {plipe} is not the best example for me.  The notion of
"jumping" naturally includes a point of takeoff, and maybe the others.
But what about {farlu}?  Is there any way to say "free falling" other
than {farlu befe zi'o bei zi'o}?  (In the sense "supported by nothing
other than air", rather than the astronautical sense, "supported by
nothing at all".)  Falling can be a much more protracted thing than
jumping usually is.

(OK, there's {se sarji le vacri}, but that seems to be somewhat
different.)

(I get the impression that you're reluctant to make changes to the
place structures at this point, but if I succeed in making a very
convincing argument it might happen.  Is this right?)

 > MANY gismu place structure decisions were made ad hoc based on specific
 > pragmatic usage considerations, and NOT on the basis of creating some
 > ideal mapping of concept space.  Indeed I think we have explicitly
 > REJECTED the idea that the gismu should be considered anything like an
 > ideal - that our vocabulary should be in some way a philosopher's
 > language combining pure essences to analytically cover all concepts.
 > Many conlangs have tried for the latter, and we didn't want to.

Absolutely agreed.  Language is a very pragmatic affair.  (And, even if
we wanted to do this, we (i.e., humans) don't have a clear enough
understanding of the semantics involved.)

 > >[1] Yes, there is a source and destination, but they're not relevant to
 > >Alice in the middle of the fall; if I recall correctly, she wonders at
 > >one point whether the fall will ever end.  That would currently have to
 > >be translated as wondering whether a terbridi has a value (!).

 > clearly la alis. za'o zo'o farlu ma
 >
 > Whether a place is relevant is less important than whether it exists.
 > You could always use lo cimni.

I suppose, though I don't really like it.

 > The tougher question is xu lo mluni cu farlu ma ma.

And the answer with the current place structure has to be {na go'i}.

 > lojbab

mu'o mi'e. dilyn.