Message-Id: <9207060508.AA10577@relay1.UU.NET>
Date:         Sat, 4 Jul 1992 00:01:48 +1000
Reply-To: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!nsn
Sender: Lojban list <cbmvax!uunet!CUVMB.BITNET!pucc.Princeton.EDU!LOJBAN>
From: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!nsn
Subject:      Clefting
To: John Cowan <cowan@SNARK.THYRSUS.COM>,
        Eric Raymond <eric@SNARK.THYRSUS.COM>,
        Eric Tiedemann <est@SNARK.THYRSUS.COM>
Status: RO

Iain asks that most FAQable of questions, "what is clefting?"

Here's how I attempted to explain it a while back:


Message-Id: <199203130330.AA09237@munagin.ee.mu.OZ.AU>
Subject: Re: Lojban Names.
Date: Fri, 13 Mar 92 13:30:12 +1000
From: nsn@mullian.ee.Mu.OZ.AU

[]

Was it you, Julian, who was asking what clefts were two months ago? OK, here's
the story. In syntactic analysis of sentences, one can often speak of a
sentence being embedded within another. Thus: "I saw that he was drowning"
is analysed (in TGish terms; go easy on me, I've just started studying these
things methodically) as: ((I)(see (he drowns))), where "he drowns" is an
independent sentence.

Now often in such anaylsis, what you seem to find is that there's an embedded
sentence, alright, but bits of it are scattered in the sentence.
"He is drowning"
"HE seems to BE DROWNING"
The analysis is: ((he is drowning) (seems)), where "seems" is a verb taking
the sentence "he is drowning" as a subject. Compare
"It seems that HE IS DROWNING"

In the first case, the sentence was split up, the subject ("HE") going before
"seems", the verb phrase ("IS DROWNING"), modified and appearing after the
phrase. This split is called clefting. In the second case, the subject and
predicate of the embedded clause are together. It's not clefted.

Now, a, shall we say, literal-minded analysis of "seems" would give the lojban
"x1 seems to be lenu x1 does x2". This is in fact still the status quo:
{mi simlu lenu mi co'a djacu morsi}. It can be argued that such a set up is
bad. For example, it is redundant: The {mi} in x1 reappears (even if usually
elided) in the x2 clause. When LLG agrees that such a clefted place structure
is redundant, it puts x2 where x1 used to be. This happened with rinka.
x1 causes x2 to be x3 by doing x4. Well, in lojban, x4 will be {lenu x1 co'e},
and x3 will be {lenu x2 co'e}. So x1 and x2 are redundant, and we're left with
the current structure, event x1 causes event x2 to happen.
"I cause you to blush by flattering you"
"The event that I flatter you causes the event that you blush"
This is simpler in Lojban than a stream of sumti, but it's not always
convenient. Thus, {le rinka} is now "a cause"; it is hard to speak now about
"a causer". {jaigau} does the trick, but I feel it fails on other grounds
(it can't do something similar to unclefted x2s).

And of course, not all place structures are uncleftable. Consider {trina}:
x1 attracts x2 by doing x3. Uncleft it: "The fact that x1 happens attracts
x2." Doesn't mean the same, does it. Now try this:

Gork attracts Dork by talking to Mork (oh, how talkative it is!)
Mork attracts Dork by being talked to be Gork (oh, what a good listener it is!)

In unclefted garb:

Gork's talking to Mork attracts Dork
Mork's being talked to by Gork attracts Dork

I don't know if it's obvious in English, but these two sentences cannot mean
different things in Lojban. Thus {trina} can not be unclefted.

Thus unclefting, which, if you think about it, can lead you to quite unnatural
ways of speaking, is justifiably a major issue in Lojban design.

Hope this made sense.
[]
---
'Dera me xhama t"e larm"e,	      T  Nick Nicholas, EE & CS, Melbourne Uni
 Dera mbas blerimit		      |         Mail: nsn@munagin.ee.mu.oz.au
 Me xhama t"e larm"e!		      |  Omiloume ellhnika/Esperanto parolata/
 Lumtunia nuk ka ngjyra tjera.'       |  {mika'e tavla baula lojban.je'uru'e}
 - Martin Camaj, _Nj"e Shp'i e Vetme_ |      (Better .sig suggestions welcome)