[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Bringing it about that
At 05:01 AM 04/13/2000 -0400, pc wrote:
The focus was on, first, getting to John, since the causal words seem to
require (quite rightly) events in both the cause and the effect places. That
meant that "John" had to be subject raised in the subject position,
Or alternatively you can amend the predicate with "jai" or even nest
predicates with la djan. gasnu lenu zukte
a
slightly odd case. And once it was entered, a deeper problem arose: if
"John" had to be raised from, say, "John's laughing made me hit him" to get
"John made me...," why doesn't "John's laughing" have to be treated as a
raising, since it is presumably something about it that worked the effect
"The fact that John's laughing was annoying made me..." And so on, ad inf.
That could be said about all causality claims, could it not?
The practical solution, that "John" needs to be flagged because it is not an
event name in a place that calls for one, seems ad hoc -- one the one hand,
John IS an event by many definitions
Yes and if his mere existence motivates something, that would seem
acceptable in x1. For my teenage daughter and certain equivalently-aged
males, this seems like a good description %^)
and, on the other hand, the fact that
"John" doesn't fit should be marker enough, whereas "John's laughing" might
be more misleading. And, of course, if we want to say that what is raised
depends upon what the real motivator was, then we have to face the fact that
that might be John (not as an event even), not something he did or was, and
so the reaising might be inappropriate, even misleading.
I think that the important point in flagging the place as an event is to
get people to think about what the real motive is before making the claim,
which usually is not just "John".
A second matter was the third place of mukti (and some other of the words in
this collection?) event x1 motivates event x2 in volition of x3. Since x2 is
presumably an even in which the agent is the one who's volition is involved,
A presumption that we explicitly wanted to avoid.
x3 repeats the subject of the description in x2, usually at least.
Agreed as to "usually", but someone may be motivated to have something done
to them. My son is having surgery done next week to remove some pins from
an earlier broken elbow. (The pain of the pins) x1 motivates (the surgical
removal of the pins) x2 with my son as x3. Of course x2 COULD be phrased
so as to make x3 the subject, but Lojban place equivalence argues against
explicitly saying that there is necessarily such a thing as a "subject".
Note that as things worked out, there is an explicit relationship between
gasnu and zukte for claiming agentive relationships either with or without
a claim of motive. mukti in a sense is a conversion of zukte, but does not
necessarily claim that the person with the motive is agentive in the action.
These
place structures are baselined and immutable until some future date, but we
might consider reconsidering these cases at that date.
There was a LOT of debate over the place structure of mukti before
baselining - it may be in the archives. There was also a change of keyword
(from "purpose" to "motive", I think it was) which was actually made after
the original baseline and took a formal decision.
They seem to arise
from English (etc., but we know what most of the founding members spoke)
Actually in this case I think we debated ourselves out of our native
language point of view - indeed we may have debated ourselves out of any
natural language point of view in surmising the possibility of an x3 who is
not a major place of x2.
expressions A motivates B to do C, which is taken to be a 3-place relation,
even though most good grammars of English (etc.) recognize it as only
2-place, the second place being an infinitive sentence "B to do C".
which is where we got the tie to gasnu/zukte, I believe.
Of
course, both pieces function separately A motivates B, A motivates Cing: the
raised subject and the suppressed subject respectively.
And indeed desire for ellipsis might cause someone to only express any two
of the three places, though zukte might handle some of the cases better.
At least, until
reconsideration, we might well drop the use of the thrid place.
And back to the original problem. I used to advocate a predicate "x1 brings
it about that event x2 by doing event x3."
Switch x2 and x3 and you have zukte.
I now notice that, except
rhetorically, this is another case of duplicating the subject of an event
description. There is, however, no obvious predicate in Lojban that does
this in the properly vague way, so that the original sentence is still
untranslatable in its full obscurity.
That vague way is indeed the intended point of zukte, which as I noted
above is parallelled with gasnu as the explicit agentive sumti-raisers.
But there does seem to be a tendency to use gasnu in something like this way
(or rather something like the original version, but without the third place).
This seems to turn up most in lujvo, where rafsi of gasnu turn up finally
whenever an activity or process is derived from a state predicate (loosely
speaking, since these are not hard concepts in Lojban).
Yes, and for every -gau predicate, there is an implicit parallel -zu'e
predicate. But only if people want to talk about the purpose or goal do
they tend to use the latter, and (thus far at least) I think people are
still prone to thinking two-placedly (there ought to be a good ten-dollar
Latinate word for that! A 10-rupnu Lojban word is easy, of course).
Even though literal
is not always best in lujvo (since not in tanru) and it is hard to make rules
about the semantics of lujvo formation, this tendency is worrisome, given the
history of, e.g., madzo (x1 makes x2 out of material x3) in Loglan, where it
spread to something very close to "bring it about that" but then also became
impersonal, beyond even English "make" and Fr. "faire" and so came to mean
very little at all (and nothing that could be traced back within Loglan to
its core meaning).
I think we have strong motivation to avoid madzo-like evolution. If only
to keep from being subject to such a criticism. Indeed, I think the
tendency is the other way, towards overly analytical semantics, especially
as compared to the poetic lujvo that Michael Helsem used to coin (and may
still, since I never have time to read his writings, sorry Michael).
lojbab
----
lojbab lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273
Artificial language Loglan/Lojban: http://www.lojban.org (newly updated!)