[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
--- In ceqli@yahoogroups.com, "oskar2379" <xeubie@h...> wrote: > I've been playing around with different loglangs and after looking > through ceqli, I am interested in how it resolves ambiguity. I am > specifically talking about figuring out what is the predicate and > what are the arguments. > > One example is: > > Pe xau kom pan sa kane. > Person sees the bread-eating dog. > > How does one know not to separate it as: > > [Pe] xau kom [pan sa kane] > Person see-eats the bread-type-of-dog. > > or > > [Pe xau] kom [pan sa kane] > Person-goer eats bread-type-of-dog. > Allow me to presume to answer before Rex does. :-) We've been working through some of these very things in the last month or so. One recent creation is the word "sa", which wraps up the preceding phrase as a unit, and is useful for marking phrase boundaries. It needn't be used where other disambiguating mechanisms are present, such as a pronoun. (In fact, "sa" is kind of a pronoun where there isn't a pronoun: "The thief, 'he' is in this room" is the sample Rex uses.) In simple declarative sentences that don't contain what we think of as dependent clauses, the need for "sa" is not as frequent. A design point of Ceqli is that speakers may omit disambiguating cues if they are truly not necessary, or if the speakers or listeners can deduce based on context, or if they simply don't care that they are ambiguous. > In both cases, the verb is confused as part of a compound. > > I'm also curious as to what you consider the advantage of putting > adjectives and adverbs and suchlike _before_ whatever they are > describing. I realize that run-on sentences are non-existent in > chinese, due to the fact that descriptions can only run on for so > long before you must reveal what you are describing. I can't > decide whether or not this is a good thing, so I'll leave it for you > all to help me with. I'll leave this one for Rex to answer. (Another ;-) ) > > oskar