Received: from access3.digex.net by nfs1.digex.net with SMTP id AA18422
  (5.67b8/IDA-1.5 for <lojbab@nfs1.digex.net>); Sat, 13 Aug 1994 04:02:50 -0400
Received: by access3.digex.net id AA02102
  (5.67b8/IDA-1.5 for lojbab); Sat, 13 Aug 1994 04:02:47 -0400
Date: Sat, 13 Aug 1994 04:02:47 -0400
From: Logical Language Group <lojbab>
Message-Id: <199408130802.AA02102@access3.digex.net>
To: jorge@PHYAST.PITT.EDU
Subject: Re: Allnoun
Cc: lojbab@access.digex.net, lojban@cuvmb.cc.columbia.edu
Status: RO
X-From-Space-Date: Sat Aug 13 04:02:53 1994
X-From-Space-Address: lojbab

JL> Could you come up with that without writing it down? Also, in your phrase
JL> it would be the table (or the table top) rather than the cheese that was
JL> left --More--
JL> by the maid, unless you want to invoke tanru ambiguity, in which case the
JL> whole sentence could mean anything.

I >DID< come up with it without writing it down, just looking at the
English and the cryptic parenthesizing that the guy on conlang who posted it
started hgenerating, and it just came out of me - (snap) - that fast.

I then wrote it down to check it.  The only really questionable part was the
maid-leaving-behind thing.

As for that, look agaoin:
> fetsyselfu pruselcliva ke jubme cpana ke'e cirla citka smacu kavbu mlatu
> jersi gerku raplydarxi nanla
>
> (female-server past-left-behind) (table upon) cheese eater mouse catcher cat
> chaser dog repeatedly-hitter boy

lo cpana is something that is >on< the table, not the table itself.  This 
is (stripping the tanru) a selcliva cpana - a left-behind type of thing which
is upon something (and this modifies 'cheese').

I find it entertaining to do observative mode sometimes, and it does start
to make some sense.  I would not be surprised to find some similar methods
used in Chinese and other languages that Americans think are 'grammar-free'
I know that whne I read glossed Chinese, it looks like observatives to me
(which is what got me first into trying to write super tanru.

I agree with you that long tanru are a pain.  UNLESS they are quite
monatonic in grouping like thsi one was.  One trick (if you are translating)
is to do just like I did and cast all words in the same part of speech.
This one was particularly easy since it used mostly gismu., and very concrete
tanru components.  My problem is with more metaphorical tanru for abstracts
like "democracy".

lojbab