Re: TECH: long, but major topic: lean lujvo and fat gismu

[Jorge LLambias <jorge@PHYAST.PITT.EDU>]

I'll first summarize the main observation I wanted to make about lujvo,
before tackling the "fat gismu" issue, which I think is not the main point.
I think the guidelines that Nick gives for the creation of gismu are very
sound in general. This is how I interpret them:
(In the following, t1, t2, t3 are the places of the tertanru, and s1, s2,
s3, .. those of the seltanru)
The _maximal_ number of places, and the usual ordering, is:

je-lujvo: t1=s1 t2 t3 t4 t5 s2 s3 s4 s5
belenu-lujvo: t1 (t2=NU seltanru) s1 s2 s3 s4 s5 t3 t4 t5
be-lujvo: t1 t2=s1 t3 t4 t5 s2 s3 s4 s5

(NOTE: In the last two cases, the place filled by the NU or by the s1,
could be t3, t4, or t5, instead of t2)

My only disagreement with that is that for be-lujvo, I think it should be:

be-lujvo: t1 (t2=s1) t3 t4 t5 s2 s3 s4 s5

i.e the place in parenthesis should not be present in the place structure
of the lujvo, because it has already been filled. This, by the way, is what
happens in most of the be-lujvo that appear in the jvoste, but it has not
been said explicitly.
This seems to give a horribly big number of places to lujvo, but this is
an illusion, because most gismu don't have five places, and there are some
other "standard" eliminations:

Eliminating places:

1- The easiest is when a place of the seltanru is filled by the tertanru
taken as an event (I've been calling this "reverse-belenu" elimination)

e.g.
balsoi (banli sonci)  s1=b1 s2 (b2=nu/ka sonci) s3

2- Often, other seltanru places merge with tertanru places.

e.g.
berstici        northwest: b1=s1 b2=s2 b3=s3

3- In a few cases, a place may be eliminated because it takes its default
value.

e.g.
bavlamdei       tomorrow: d1=l1=b1 l2=b2 (d2=1, default) d3
(I don't like much this type of elimination. Why not leave the place
with its default, so that normally it will mean "the following day", but
we could use it for "the following three days" if needed?)


In all these cases, and I'm sure there must be some other possibilities, the
places eliminated are clearly being filled by something, so I have no major
difficulty in accepting that they are not part of the lujvo.

What I find harder to accept, is the elimination "because it doesn't fit
the concept I want". Nick says that this is often an indication of the
wrong gismu having been chosen, but accepts it in some of the examples.

Here is where "fat gismu" come in. It is true that sometimes, it seems
wrong to leave a place, but there doesn't seem to be a better gismu to do
the job, so that a gismu+zi'o seem to be required. What I say is that if
such a gismu is required to form a lujvo, then that concept is a valid
concept by itself, and the gismu is "fat". I don't think there are that
many gismu in this situation, but I'm sure there are a few.

Whether {gerku} is fat or not is a matter of opinion. It wouldn't make
much sense to eliminate its x2 because all animals have one. On the other
hand, the concept without the x2 does make sense, as does {remna}.
I don't think it's worth even cosidering dropping it, but we should be
aware that it is this (slightly) different concept that we are using to
form the lujvo {gerzda}, and I think it's a nonstandard formation.

To lojbab's specific points:

> Moreover, neither of these expresses what we USUALLY want to say for a
> doghouse, which is that
>
> x1 is a gerzda (house-for-dogs) of-dog (specific) x2

This is true only if you give dogs their preferred domestic animal status,
which gerku by itself does not suggest. You wouldn't want {mantyzda} to mean

x1 is an ant-nest of ant (specific) x2,

but rather

x1 is an ant-nest of ants of species x2.

I think this is the general case, and if you want the other for dogs, it's
only because we tend to think of them more as individuals, than as just
a member of their species.

> Thus the person who proposes the lujvo "gerkyzdani" has to decide just
> which concept they want to represent:
>
> x1 is a doghouse of/for dogs x2 of species/breed x3
> x1 is a house of/for dogs x2
> x1 is a doghouse of/for dogs of species/breed x2
> x1 is a house of/for dogs

Actually, my main argument was that the first two are not good options,
and should be eliminated from the start as possibilities. This is because
the x2 place of {zdani} is already filled, so it doesn't make much sense to
me to allow it to be filled again.

I agree that the third does not seem very useful, but I can't see any
general principle by which to eliminate the breed. That's why I say
that what is modifying {zdani} is not {gerku}, but {gerku be zi'o}.

I'm definitely not suggesting that the {zi'o} be made explicit in the
lujvo.

I don't think {mensi} (sister) suffers from this problem. Here
{mensi be zi'o} is pretty meaningless, or at least quite useless. It is
not what I would call a fat gismu.

> Returning now to argument #1, my answer is that you CANNOT
> algorithmically exactly which places gerkyzdani "needs".  All of the
> possibilities mentioned above are distinct and plausible interpretations
> of the tanru "gerku zdani".

To me, only the third one is the "standard". I'm not saying that lujvo
should always follow these rules. Of course there will be deviations that
come from usage. My point is that there can be ground rules, which will be
followed by the great majority of lujvo (as they seem to be) and those
that don't follow them will be exceptions, as in the case of {gerzda}, if
it comes to mean what is proposed, due in this case to the special
relationship we have with this beast.

>  The one that you choose as the place
> structure for gerkyzdani ought to be the one that pragmatically turns
> out to be most useful.

And who determines this? In this case, it may be more or less clear what
pragmatics suggests, but in others it may be more doubtful. I think that
deviations should be left to usage, rather than being introduced from
the design stage.

> If I have misrepresented the arguments against fat gismu, I'm sure I will
> be corrected %^).

Well, {gerku} is not the best example, although it was the one that came
up from the lujvo paper.

> Similarly, if "I go to France" and "you go to France", we can say "we go
> to France" without implying that we started from the identical origina,
> used the identical route, as well as the identical means.

Are you saying that {mi e do klama le frasygugde} means the same as
{mi'o klama le frasygugde}? I think the latter means that we start
from the same origin, same route and means, is this wrong?

> Nick has proposed some 5 or 6 tanru
> interpretation schemes that are MOST COMMONLY the basis for tanru that
> are in turn the most useful basis for lujvo.

3 if I'm not greatly mistaken. (Actually 2, since belenu- is a special
case of be-lujvo.)

> But I would contend that the determination of place deletion once you
> have selected one particular scheme needs a totally separate analysis
> (which Nick has more or less discussed in his paper, but I get the
> impression from the followup discussions that it is not treated as a
> totally separate problem).

You're right that we've been mixing the issues. Hopefully they're becoming
clearer as we discuss.

> Which should be chosen for the dictionary?  A little easier question,
> but not much.  First of all, we have to make it ABSOLUTELY CLEAR that
> all lujvo place structures at this point are ONLY proposals, and that
> actual usage may cause them to (need to) be changed.  This is no
> different than what I say for the gismu place structures and why I
> refuse to baseline them even when the dictionary is published.

A very commendable attitude. This (hopefully) could help to get rid of
the fat gismu.

> So what to propose?  I suggest that the propensity to use lujvo as
> term-bases for longer lujvo should lead us to eliminate places where
> practical, i.e. make our lujvo relatively "lean".

I agree, as long as the eliminations do not require the underlying gismu
to change their meaning.

> We don't want
> "gerkyzdanydinjyzbasu" (dog-house-building-maker) to have all of the
> possible places suggested by the components:
>
> gerku ger ge'u dog x1 is a dog/canine of species/breed x2 ai 175 (cf.
> lorxu, labno, mlatu)
>
> zdani zda nest 'den' x1 is a nest/house/lair/den of/for x2 4d 280 (cf.
> dinju, ginka, kumfa, xabju)
>
> dinju dij di'u building x1 is a building/edifice for purpose x2 2k 153
> (cf. ginka, zdani)
>
> zbasu zba make x1 makes/assembles/builds/manufactures/creates x2 out of
> materials x3 7f 217 (cf. cupra, larcu, rutni, finti, gundi)
>
> So I propose
>
> gerzdazba
> x1 is a builds doghousebuildings (for houses, dogs of breed, purposes)
> x2 out of materials x3

This is what I propose, following the rules given above:

z1 (z2=d1=z1) z3 (d2=NU gerzda) (z2=g1) (g2?)

x1 is a builder of doghouses of material x2

(Even leaner than the one you get. I cheated in eliminating the breed,
though) I don't think you should keep the place for the built object,
because you already know they're doghouses. If you want to give more
details, then you can just use {dinju}.

> Not always will the place structure match the final term.  An example
> discussed in commentary was
>
> ponsydjica own-desire
>
> ponse pos po'e possess x1 possesses/owns x2 under law/custom x3
> (cf. ckini, ralte, jitro, steci, srana, tutra, turni)
>
> djica dji desire x1 desires/wants/wishes x2 (event/state) for purpose x3
> *3l 500 [if desire is for an object, use tu'a in x2]; (cf. taske, xagji,
> nitcu, nelci, pacna, prami, rigni, trina, xebni)
>
> for which I suggest
>
> x1 desires that (x1) own x2 under law/custom x3
>
> The law/custom and the thing owned comes from ponse, and not from djica,
> and I have dropped the purpose place of djica.

This fits the rules, although it's the first example I see of a
double-belenu. i.e., both the desired event, and the purpose of the
desire, are {le nu ponse}.

> I tend to like to delete places when the option exists to add back in
> (relatively unambiguously) using a BAI tag.

If only this were true for some gismu! For example, the {bau} places, which
are usually inaccesible without a FA anyway, where a {bau} is so much clearer.

> I will always delete a
> place if the abstract concept I have in mind doesn't metaphysically
> require the place.

Ah! this is a totally different matter. If you start making metaphorical
lujvo, almost anything can happen.

> This is long enough.  Let's let people start shooting me down.  Jorge??? %^)

With pleasure. :)

Jorge