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

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

Re: [engelang] Xorban: la je cmla nltra


On Mon, Oct 22, 2012 at 10:20 PM, John E Clifford <kali9putra@hidden.email> wrote:

From: Mike S. <maikxlx@gmail.com>

I think predicates for the deictics are desirable, but I want to think more about "dz" before I reply here.

But deictics are among the least likely things to have predicates for them in natural languages; they are regularly "pronouns".  They are pragmatically defined (like personal pronouns -- at least first and second) and have no semantic content per se.

Somewhat true re deicitic predicates, though you do have interesting periphrasis like "He is I".  In Xorban, 90%+ of the time, when a speaker wants to say "me", he'll say "a'a", and "a'a" isn't a predicate.  "la'a mba'a" is usually the implicit binding of the first person entity, but it can be made explicit when you really want to emphasize "me" (like Japanese "watashi").  I find your suggestion that pragmatically defined things "have semantic no content per se" incomprehensible.  Of course "me" has semantic content -- it refers exactly to the person uttering the sentence.  That's perfectly meaningful.

There are other deictics that don't have variables but only predicates in X -- "tj, tz, tf" = "this, that, that over there".  These also make sense.


 
I don't think that building L "ka" into an argument place builds significant stuff into it, because I don't think that "ka" signifies much at all.  "Lo broda" and "lo ka broda" are essentially two interchangeable ways of looking at the same thing, and although Lojban has this distinction all over the place, it's trivial to switch back and forth: "lo ckaji be lo ka broda" = "lo broda" and "lo se ckaji be lo broda" = "lo ka broda", or am I mistaken?

The fact that one can shift expressions around in this was does not mean that something different is not involved.  The sorts of things that  characterize are not the same sorts of things that they characterize.  Rats are concrete physical objects, ratness is not any of those things, and ratnessness is even less so.  There are connections (and just what those are has driven philosophers to various kinds of madness for two and a half millennia).  You can say pretty much anything you want to say about rats in terms of ratness, though it looks pretty strange.  It is harder to say things about ratness (either kind) in terms of rats.  A good case can almost be made for just doing without ratness (etc.), but, alas, in some intensional contexts, only ratness seems to do the sort of work needed.  To be sure, we can find dodges around actually saying "ratness" or some equivalent term, but we end up using the notion all the same -- just hiding the fact behind idioms of one sort or another -- which will have to be decoded in the semantics anyhow.

I would get rid of "ratness" if nothing else seemed to work.  And let's not even speak of "ratnessness", which I would burn with fire, because it has no useful meaning for human speakers nor even for human philosophers, and therefore bears no consideration in the engineering of a speakable loglang.   Montague realized that intensions of intentions are all constants, and therefore useless in his program.  It's elephants all the way down, when it comes to "higher order intensions/properties".  Plato would have probably said the same thing;  he didn't talk of forms of forms, just of forms. Right?  One degree of abstraction, if needed at all, is more than enough.  And often it isn't needed.  Often "ka" is not needed.  So why force it on Xorban.

--
co ma'a mke

Xorban blog: Xorban.wordpress.com
My LL blog: Loglang.wordpress.com