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

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

Re: [engelang] Xorban: Properties


I'm just recording a few preliminary thoughts to be pursued at a later time...


On Tue, Oct 23, 2012 at 10:33 PM, John E Clifford <kali9putra@hidden.email> wrote:
From: Mike S. <maikxlx@gmail.com>

la fgra le glre dksake.
versus
la fgra le li glri kmeki dksake.

It is you, not I that have created these monsters of imprecision, "The fire is extreme among the hot things", e.g., not me.  Saying that it is an extreme in a scalar property (or, indeed, as Xorxes points out, beyond a certain limit -- though the predicate doesn't say that) is certainly better than that (and btw would just have a predicate maker, not this strange predicate) .  One would like to just say that it is too hot -- adverb-adjective, but you have set things up to make that impossible, just as we can talk about a green house in any ordinary way.  These kinds of primitive bad decisions are the sort of thing that should get caught early and corrected, not defended in the face of obvious difficulties.  God help us if we try to change the phonology.

I thought that "dks" worked as intended, but how would you handle "The fire is too hot for x3"?

Richard's properties (modal) are useful for many things, but are not the last word.  As normal cases like "If rats were insects" show, the extension of "rat" in some worlds may contain no rats(as we understand them), which hardly seems right for properties.  (A consequence of this mechanism is Goodman's "This is the best of all possible worlds because it *is* all possible worlds") Plato is not a very good case, since, as Aristotle and others  before him noted, his system is classically incomplete and entails (unacknowledged, of course) that there is at least a Form of forms?  Properties are often not necessary, but also often are.  And remember that, given properties, individuals can be eliminated -- but not conversely.
I can't say too much about what a form of forms might be, but I can say that in Montague Semantics, a property is simply a function from worlds to a one-place predicates, which is covered by X "l-" as long as intensional readings are admitted.  I believe that "km" would be either redundant or an encumbrance in ordinary Xorban.

My point is just that Richard, dear man that he was, is not Frege nor Church and his notion of properties is defective in just that way.  They work for a lot of things, they don't work for a lot of others.  I am not clear what you mean by intensional readings of l but I have come to expect some strange sort of misuse of at least terminology, if not of fact,  when you talk this way.   Do read up a bit, please, this stuff has been worked over for years.

By "intensional reading" I mean that the predicate place is ranging over the connotations of the supplied argument, rather than over the denotations. 

I read the "properties" article (interesting reading) and related over at SEP,

http://plato.stanford.edu/entries/properties/

and what I am seeing right now that we already have two or three ways to get at properties:

1. Predication.  "la fgra glra" means "The fire is hot" and could also be said to mean "The fire instantiates hotness".

2. Copula.  "la fgra le glre vsmlake" means "The fire visually appears hot" and "The fire visually appears to instantiate hotness." 

If (1) is analyzed as "la fgra le glre mnake" = "The fire is among hot things" or better yet, "la fgra le glre mplake" = "The fire instantiates something hot", then all copula-like constructions, i.e. all those invoking the instantiation of a property, would have similar syntax.  "dks" from above meaning "too much" is also in this class IMHO.

3. Situations. "le fe la fgra glra plkeka'a" means "It's pleasant to me that the fire is hot" which can be viewed as a crude approximation of "The hotness of the fire is pleasant to me."

I don't see any problems with the copula-like property-instantiations of 1&2.  In these cases properties occur transparently in the "l-" construction or predication, and it would be redundant and needlessly verbose to use extra syntax to call special attention to them.  I think "dks" works fine.

However (3) leaves something to be desired.  Literally it says that the situation in which the fire is hot is pleasant, not that the hotness in particular is pleasant.  A few possible solutions:

3a. Focus.  "le fe la fgra ni glra plkeka'a" = "It's pleasant to me that the fire is HOT".   We have seen that "ni" works as focus marker. This solution pushes property references into the realm of the information structure, but it's very versatile and may do the trick in some cases.

3b. Periphrasis.  le je li fgri mplike glreki plkeka'a.  "The hot thing instantiated by the fire is pleasant to me" makes it clear that it's the hotness that is pleasant.

3c. Operators.  I don't have any proposals yet, because the common logical form of properties seems uncertain, the semantics are blurry, and there are many kinds of properties (e.g. some are self-instantiating and some are not; see SEP article on properties, and several related articles, for other examples).  I will only point out that importing "ka" will not work any miracles for us, because "ka" itself doesn't seem to do everything that we need it to do, or at least not too easily.  The "km" predicate doesn't make it very easy to say "the fire's hotness" for example, and, unless I am mistaken, neither does "ka", because the "ce'u" blocks where "lo fagri" would go.  These are sometimes called "tropes" but are better called "properties of particulars".  There are also relations and the determinable/determinate distinction to consider.  If we create something for Xorban then we're going to want it to be both simple and comprehensive, and I don't see that happening until someone creates a lexical taxonomy of properties and the constructions that they can syntactically appear in.

--
co ma'a mke

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