Re: [lojban] Lions and levels and the like

[maikxlx <maikxlx@gmail.com>]

coi.  I am going to practice my Lojban -- please stand back!  Corrections welcome.

On Mon, Nov 14, 2011 at 10:26 PM, Pierre Abbat <phma@phma.optus.nu> wrote:

Would it be correct to say "lo'e za'e grezunca'a cu se finti la .caklis."?

i mi na jinvi. .i ba'e lo za'e grezunca'a cu se finti la .caklis. .iku'i no da poi za'e grezunca'a zo'u da se finti la .caklis. .i ku'inai naku lo'e za'e grezunca'a cu se finti

 

Also, does my example "lo'e .ornitorinku na fadni mabru .iki'ubo na'o se
jbena re sovda" involve a kind?

i frica.  .i lo .ornitorinku na fadni mabru .ije ja'a ro da poi .ornitorinku zo'u da na fadni mabru .i ku'inai ja'a lo'e .ornitorinku na fadni mabru

It may seem strange that given the slots of two bridi satisfied by "kinds", no individual satisfies one slot, and all individuals satisfy the other slot.  Part of the difference, in my best analysis (very much ongoing), is that there are really (at least) two sorts of "kinds" or more accurately two sorts of "kind" predications, which we can call (true) kinds, and generics, each of which go by their own logic.  In the case of the (true) kind, the predicate is episodic i.e. refers to one event or situation:

(1a) Trees are widespread.  [trees as one giant mass, one "snapshot" glimpse]
(1b) Dodos are extinct.  [dodos as a one species, one extinction event]
(1c) Birds evolved from dinosaurs. [(at least according to science) similar to (1b)]
(1d) Transistors were invented by Shockley et al. [the whole class of components, one invention event]

In none of these cases does it seem that ANY individual satisfies the predicate that the kind does, at least not with the same reading.  Generics work differently:

(2a) Dogs are mammals.  [intrinsic property for individuals: All(x): Px -> Qx]

(2b) Insects are six-legged.  [characteristic near-universal property: Most(x): Px -> Qx]

(2c) Lions are ferocious.  [characteristic common property: Many(x): Px -> Qx]
(2d) Humans have walked on the moon. [existential property: Exists(x): Px and Qx]

I annotated each case using (non)standard quantifiers with "x" as a variable of individuals, Px corresponding to the English subject NP and Qx to the VP, in order to compare what portion of the individuals satisfy the predicate .  In these cases, it seems that some individuals at least MAY satisfy the same predicate that the generic does.  In fact, generic interpretations seem to arise "squinting" at the whole class.  More precisely, we are quantifying over individuals and situations.  

I believe that the reason for the difference (assuming what I am saying here holds up) basically boils down to VERB ASPECT.  Kinds seem to involve a perfect(ive) aspect of some sort describing episodic events that don't pertain to any individual.  Generics seem to involve some sort of generic, habitual, or similar aspect describing numerous situations that may involve every individual.  In Lojban, these two very different aspects need not be marked.  However, by avoiding the need to mark aspect or any other grammatical category (which works fine in normal human usage I think), Lojban does incur a little bit of confusion in the formal semantic analysis.

Incidentally, IMO (as shown above) given the canonical definition of the "typical that really is", {lo'e} can refer to generics but not kinds.  If the "typical" broda does something, then certainly at least one individual broada should be able to do it, which true kind-predication disallows.  One can argue that {lo'e} thus can mark generics at least, but I think it would make more sense to use aspect cmavo on the bridi.  What boils down to verb aspect shouldn't be marked on the gadri.  That's totally my 2 cents though.

Pierre

-Mike

-- 
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.