[lojban] A summary on 'djica' etc.

["John E. Clifford" <kali9putra@yahoo.com>]

1. There are no grammatical restrictions on what kind of term can occur in an argument place in Lojban.  Some may seem more natural in some places, some may seem idiotic in some places, but all are grammatically allowed.  The natural and idiotic features of some kinds of terms in some places is largely a semantic matter: some predicates just seem to beg for solid objects in various places, other seem to only be comfortable with propositions, say.  The source of the naturalness or not may be found by examining the inner logic (the ultimate semantic deconstruction) of the predicates and terms, although this is often hard to do in final details. and in the process we may discover a way that terms not initially thought to go in a slot might nonetheless go there and an explanation of why they make sense there.

2,  There is in Lojban, as in many languages, a grammatical process called raising, by which a term in a subordinate position is brought into a higher clause.  It may come to replace that subordinate clause or to fill another position in superordinate clause.  This is a sort of reverse of the process of eliding information that is repeated, when it is "obvious". So, in "I'm thinking of buying a car" we don't mention that it is me buying the car; that's obvious.  But similarly, I might report those same thought as "I am thinking about a car", raising "a car" from the clause "that I buy a car" (in there at some level) to replace it.  "I said "He's an asaph" about Tom (not Peter as someone just claimed), raising the identification "'he' = Tom" to the surface from the quote and the pragmatics of the situation.  
This process can always be carried out, grammatically (well, I  suppose each language has some limitations), but logically it is a risky business, particularly when the clause from which it is raised is a world creating one.  A world creating clause is one that invokes an alternate story to the one in which the main line develops.  Typical examples are modal claims (what is possible or probable), contrary-to-fact conditionals and other subjunctive clauses, and remote tenses, especially the future.  The problem is, that in these alternate stories, there may be different things and even familiar things may behave differently, so in moving a term out of a world where it has a certain referent into the current world, it may lose its referent (or introduce a new thing, its referent, into the current world) and thus turn a true statement into a false one -- always a no-no in Logic.
Not really incidentally, the reverse move, of putting reference from the main story into the alternate world is usually not problematic, since -- except in extreme cases -- most of the of the main story carries over into subjunctive and possible and even future worlds.

3.  At the core of the logic of predicates of desire is a description of how that desire is to be satisfied.   This tends to be put in terms of a felt lack or gap (whether is is phenomenologically justified talk or not I don't know, do we introspect a feeling of a gap?) and what would fill it. That is we have a counterfactual, or, rather, a pair of them: "If such and such would happen then this gap would be filled" and "If this gap were filled, then such and such would have happened".  These two define a desire (want, need, wish,...) for such and such, clearly an event of some kind.  So, the natural term after "I want" is an event-description of some sort.  

4.  But such event descriptions often feature a non-event term as the focus of the desired event and, indeed, the event desired may be very indistinct compared to the clarity of the focal object. 
So, it seems inevitable that that object term be raised to the position of the object of desire: I want to eat an apple >  I want an apple to eat > I want an apple.  

5.  But now the problems with raising arise: 'an apple' which had its referent in two alternate worlds, now appears to have it in this one.  And it may not.  We need a particular apple which is the one I desire.  Clearly, any apple will do for the first part of the definition, If I eat it (say), then my gap is filled.  But not the second part, since, in those remote stories, there is no guarantee that my gap was filled by this particular apple -- indeed that this apple is even in that remote story.  So, we have gone from the true claim that I want to eat an apple, to the false one, strictly speaking, that I want some particular apple.  Of course, this problem does not arise in English, which has the occasional virtue of being sloppy: we say "Oh, you know he still means that he wants to eat an apple."  But Lojban does not have that virtue -- or at least is not supposed to.  If I say 'mi djica lo plise', I can't claim that I really said 'mi djica lo nu
 citka lo plise'.  
Of course, this problem does not arise if you pick your apple before hand, as it were, in this story before going to the alternates.  So, 'mi djica le plise' is not a problem, since that is a raising from 'mi djica lo nu citka le plise' which carries its referent back to the alternate worlds and is the same in both of them (the referent of 'lo plise' need not be, which makes the raising doubly suspect in this case),

6. So, to keep the logical part of the logical language going, so that grammatical transformations do not involve invalid arguments (moving from true to false), don't do raisings out of alternate stories.  To meet the urge to say, a la English, "I want an apple", leave at  least the vestiges of the event description (in these case; perhaps other things in other cases) in place. Or pick your desideratum beforehand;  'mi djica ta' is not going to giver you problems.


      

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.