place structure of lujvo

[David Cortesi <cbmvax!uunet!UUNET.UU.NET!pucc.PRINCETON.EDU!infmx!godzilla!cortesi>]

There's a recognized phenomenon in early language acquisition
called "overgeneralizing," when a child first figures out a rule
and then overapplies it.  Picture the frustration of a 4?-year-old
who has just discovered that the plural of "man" is not after
all, "mans".

Sometimes I feel like that child w.r.t. Lojban.

What brings on the sensation most recently is a correspondence with
Nick regarding his translation of Aesop.  Brutally shortening,
it goes like this:

Nick (in story):  le maxri lei manti se sudri'a

Me: I read this as "the wheat the ants dryly caused"

Nick: The x1 of sudga becomes the x2 of sudri'a...[further explanation,
        leading to]...sudga means is-dry;
        sudri'a is a single word meaning to dry: x1 dries x2 of x3

Me: You are saying that for gismu G and H, the argument pattern of a
        lujvo GH is:  h1 GH g1... ?

Nick: Actually, I was going with h1 g1 g2 ... h2 ...

Me: I thought the arguments of a lujvo GH were just those of H.

Nick: that seems to me an excessive restriction, which makes lujvo a
        very limited special case of tanru.

Me: [subsides in bafflement]

OK, I recognize that a lujvo is different from the tanru made from
the same components. (I seem to recall Bob L. saying that it would
be more specific, having only one conventional meaning out of all
the possible interpretations to which a tanru is subject.)

What I guess I don't understand is the range of possible differences.
In particular, I am (like the 4-year-old) desperately looking for rules
that I can apply.  When I meet a strange lujvo, like Nick's "sudri'a,"
under what rules do I translate it?  If I cannot simply unpack it to
make a tanru and translate that, what am I supposed to do?