Equivocal tanru

8 views
Skip to first unread message

Escape Landsome

unread,
Mar 17, 2012, 10:04:14 AM3/17/12
to loj...@googlegroups.com
I asked some question which was not understood, so I ask it anew :

a tanru X Y is some Y that has X-ness. it is not "X and Y" (unless
we play with the predicativity of X and Y), rather it is one Y, but
with X semantic coloration.

If I want to speak about a X Y Z, that is, either a Z which is
somewhat like a X Y (in the tanru sense above), or a Y Z that is like
a X, --- but keeping myself deliberately equivocal, how do I proceed ?

The way Lojban does seems to compel me to disambiguate, but what if I
WANT TO REMAIN AMBIGUOUS ?

How do I say "little girls' school" in lojban WHEN I WANT TO REMAIN AMBIGUOUS ?

--

Please consider not this as a mere amusement of myself... If we want
to translate old books of yore in Lojban, we'll have to remain
equivocal with some "X's Y's Z" phrases, BECAUSE the books in which
they are written are equivocal themselves (and we ought not to
desambiguate things that are ambiguous).

gleki

unread,
Mar 17, 2012, 11:46:12 AM3/17/12
to loj...@googlegroups.com
What texts are you gonna translate ? Are those texts in English ?
If you want to translate 
"quick brown fox" why not say {bunre je sutra lorxu} ?
It will mean 
brown=>fox<=quick
whereby you will show that F (fox) is both semantically B (brown)-colored and Q (quick)-colored.
But the relation between F and Q will remain unknown.

John E Clifford

unread,
Mar 17, 2012, 11:59:15 AM3/17/12
to loj...@googlegroups.com
Well, you can't be ambiguous in one sense (the one I take you to mean), because Lojban has default grouping to the right, so XYZ is automatically X(YZ), so If you want to say something that might be either "a pretty school for girls" and "a school for pretty girls", you can't, officially. You have, unofficially, two outs.  1) The nature of the bond between modifier and modified  is unspecified, so it can almost as easily be that X characterizes all the Y in Z as X characterizes Z (insofar as) containing Y.  2) People are sloppy, even in logic, so XYZ may be understood, despite what is said, as (XY)Z, based on context or inflection or whatever.  So the unmarked form is at least pragmatically ambiguous (standard logic game: until we have a thoroughly developed pragmatics, we can push all our problem case off on it, cf. God).


From: Escape Landsome <esca...@gmail.com>
To: loj...@googlegroups.com
Sent: Saturday, March 17, 2012 9:04 AM
Subject: [lojban] Equivocal tanru
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsub...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.



MorphemeAddict

unread,
Mar 17, 2012, 1:11:18 PM3/17/12
to loj...@googlegroups.com
Or you can explicitly state all the various ways it's ambiguous. 

stevo

To unsubscribe from this group, send email to lojban+un...@googlegroups.com.

Escape Landsome

unread,
Mar 17, 2012, 1:20:03 PM3/17/12
to loj...@googlegroups.com
2012/3/17 MorphemeAddict <lyt...@gmail.com>:

> Or you can explicitly state all the various ways it's ambiguous.


This is not a *very good* solution, for ambiguity is often very wide.

For instance, "three men ate four cakes", has a lot of possible
meanings... and I'd rather not list all of them.

Beyond this simple remark, there's the notion that structural
ambiguities come as continuums, whereas more superficial ambiguities
often can be listed. But (X Y Z) = perhaps ((X Y) Z) and perhaps (X
(Y Z)) [and even perhaps (X ... Z) * Y] is a kind of ambiguity that
may seem to be superficial, on a logical-syntactic level, but is a
structural deep one, on semantic or pragmatic level [e.g. in poetry]

Reply all
Reply to author
Forward
0 new messages