Re: [lojban] The efficacy of Lojban's grammar.

[Stela Selckiku <selckiku@gmail.com>]

On Mon, Apr 12, 2010 at 2:21 PM, Robin Lee Powell
<rlpowell@digitalkingdom.org> wrote:
>
> I was asserting that people actually *do* use the formal grammar
> internally, and you're asserting you don't.


I think we must be saying the same thing in different ways.  My guess
is that our internal experience of parsing Lojban is the same.  I do
parse it according to the formal grammar (as best I know it, which is
not quite to the level of the official insane mess) but I do so mostly
habitually, through recognizing large forms and bringing them into the
parse tree wholesale, not by parsing each word separately.  I'm sure
that you can read simple sentences at a glance, and I'm sure that on
some level you do so through pattern matching, since that's the only
way to get up to a reasonable speed.

For instance if I see "lo broda lo brode cu brodi" obviously the parse
tree I create in my mind is a perfectly standard one where the "lo
broda" goes in the x1 and the "lo brode" goes in the x2 and the selbri
is "brodi".  But I sure as hell don't go through that sentence word by
word, slowly elaborating the parse.  As soon as my eyes have vaguely
fixated on the sentence I can see that it goes "lo (something) lo
(something) cu (something)", and by then my internal parse has already
imported that whole structure that I'm terribly used to.  Next I go
about reading what the actual brivla are and putting some meaning into
the thing, hanging some flesh on the bones.

To put it another way, the grammar that I find myself using is not
actually the formal grammar, but a vastly elaborated equivalent one,
for convenience.  It has numerous terminal nodes with complicated
shapes like "gadri brivla/tanru prosumti NU bridi" or "gadri NU bridi
cu brivla/tanru", all applied of course after washing off the UI,
to/toi, etc.  Then I can transform that structure on demand into the
detailed perspective.

To put it yet another way, when I read a sentence that goes "lo broda
cu brode lo nu lo brodi cu brodo lo brodu", I swallow it in two bites:
I get everything up to the nu, creating a partial parse in my mind
with a big blank section for whatever's in the nu, so I've got "lo
broda cu brode lo nu ((some bridi))".  Then in the next bite I'll
process the bridi after the nu, just like I'd swallow a short sentence
whole, and plug that parse into the open slot in the first bite.  Now
I'm considering the meaning of the whole sentence, but the meaning is
emerging from the interface of two unofficial chunks, not from an
internal parse that superficially resembles the official grammar.

Subjectively I find that I'm not only able to understand and parse to
a formal grammar like Lojban's, but to do so quite easily and
habitually by switching constantly between numerous functionally
identical grammars in my mind, depending mostly on how much I'm able
to pattern match before having to get down into the weeds.

Does your internal process really seem different to you than that, doi
.camgusmis.?  Or are we just having a miscommunication?  I'm curious.

mi'e la stela selckiku
mu'o

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.