[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >

[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [engelang] intensions & extensions (Xorban)

On Thu, Sep 13, 2012 at 7:18 PM, And Rosta <and.rosta@hidden.email> wrote:
 

Mike S., On 13/09/2012 23:38:


> On Thu, Sep 13, 2012 at 3:31 PM, And Rosta <and.rosta@hidden.email <mailto:and.rosta@hidden.email>> wrote:
> 3. Split f- into intensional (world-shifting) and nonintensional versions, say f- and h-. It can be a bit verbose, tho: "la fa la sma pvjrna pxro'ekaka'a" as opposed to the ambiguous (or extensional) "la pvjrna pxro'ekaka'a".
>
> If "l-" were intensional then I think the f- and h- distinction would
> be unnecessary. I think we need a predicate or operator that somehow
> means "imaginary/possible but nonexistent in the world of the
> containing formula" which I think is what you're after, but I am
> unsure how to derive that.

"imaginary/possible but *not necessarily existent* in the world of the containing formula" -- I want to be able to talk about drawing a picture of a T-Rex fighting a unicorn without either claiming or denying that the fighting exists in the same world as the drawing.

I think that's exactly what you get with ordinary intension.  If "l-" had an intensional meaning then

lu fu sa pvjrna se trnsrse dmbake si pxriku vska'eki

would mean "we see a picture of (an intensional state of affairs of) a unicorn fighting a T-rex" with no claim of the existence of the fight in "our" world one way or another.  I think sa/se could be either le/le or sa/se without any difference.  Either way, they would inherit whatever world that "dmbake" inherited from "fu", which ultimately gets assigned by "pxriku", simply because intensional readings with "dmbake" are pointless.   In fact they are pointless in many predicate places, so like "dmbake", many predicates would compel an extensional reading on its arguments, usually wrt to the world of its own state of affairs.  In general, for each argument place, a predicate could unpack and use an intension in whatever way best suits its meaning, which does add some complexity to predicate design.
 
And of course I also want to be able to talk about drawing a picture of a T-Rex fighting a unicorn and claim that the fighting exists in the same world as the drawing.

su fu sa pvjrna se trnsrse dmbake si pxriku vska'eki
 

The problem with relying on intensional l is that the distinction between (i) there being an explosion and me looking at it and sketching it and (ii) me drawing a picture of T-Rex fighting unicorn would -- if I understand your idea correctly -- be based on "sa fa" for (i) and "la fa" for (ii), but in that case we would not be able to apply singularization to (i) without simultaneously giving it interpretation (ii). By contrast, I propose "sa/la ha" for (i) and "sa/la fa" for (ii).

I'm not wedded to h-, but I do think the (i)--(ii) distinction is necessary and also distinct from singularization.

Yes, that's quite right.  But I was thinking, well, the intension *is* already a singularization, so why not use it and kill two birds with one stone? If you need to signify extensionality explicitly in those cases where it the predicate isn't going to compel it anyway, then maybe one should suck it up for once and deal with "s/r" scope issues.  Otherwise, why not just make everything singularizing, and add particles when you want scope to matter?  I have thought about that, as I still occasionally make s/r scope mistakes.  The intensional "l-" solution could potentially work with the basic triple binder system currently in place.