No Subject

[lojbab (Bob LeChevalier)]

Subject: A second way to handle cleft place structures

2nd of 6 related messages

There is a second way to handle cleft place structure situations in a
Loglan/Lojban with uncleft structures.  (Note that all versions of
Loglan are plagued by cleft structures, and that in Lojban we are merely
recognizing and trying to resolve the semantics problems that result.  I
welcome Steve Rice or any other Institute Loglanist to address the issue
and indicate if the Institute has recognized or resolved it (and how)).

This other method is to use "gau" from selma'o BAI.  One of the few
place structure changes that have been formally adopted is the
clarification that gasnu means "x1 is the actor/agent in doing
event/process/activity x2".  In other words, "gasnu" is inherently
defined to be cleft, with the actor/agent extracted from the action.
There is no meaning to "do" that avoids a cleft structure.  "lifri" is
the corresponding (also cleft) word for a passive/patient/experiencer.
("zukte" is a third word with a mandatory cleft structure since an
action with goal requires an 'actor' to adopt that goal.  We may find
that a couple more words must have cleft structures due to the inherent
mental state of an actor that must be identified to evaluate the truth of
the predication.)

Using John Cowan's example from message 1:

            lenu          mi cinta cu galfi    le  bitmu
            The event-of (I  paint)   modifies the wall

is expressed using:

            gau        mi                             galfi    le  bitmu
            with-agent me (some-x1-event-unspecified) modifies the wall

This gives the same effect as sumti raising, but is more clear in that
sumti raising need not always involve raising an 'agent' (e.g the door
that is tried).  It also, by avoiding the difficult place in the place
structure, weakens that place structure, and more importantly fails to
recognize a hidden logical structure.  Indeed, there is no explicitly
marked link between the added agentive place, and the ellipsized x1
sumti.  In causality discussions, for example, a claim about the agent
of a change may be independent of the event that physically causes,
motivates, justifies, or logically entails the result. sumti-raising is
more vague semantically, but more precise logically.

Put another way BAI clearly specifies the semantic relation between the
sumti and the rest of the bridi, while labelled sumti-raising clearly
specifies that hidden ellipsis is present.  Both methods are a kind of
ellipsis, and both have a role in the language.  But let it be
recognized that only explicit elucidation of the hidden ellipsis gives a
complete statement, just as explicit elucidation of all places in a
bridi place structure makes a bridi more complete.

This proposal was the first solution we came up with, and in analyzing
it, we realized that it entailed other requirements in the language,
including the only grammar change in the entire proposed solution.  That
change is discussed in message 3.

----
lojbab = Bob LeChevalier, President, The Logical Language Group, Inc.
         2904 Beau Lane, Fairfax VA 22031-1303 USA
         703-385-0273
         lojbab@snark.thyrsus.com