In a message dated 1/31/2002 9:27:26 AM Central Standard Time, pycyn@aol.com writes:{sei} outside the homeground of inserted attributins in quotes. Oops! that is {sa'a}, this is {sei}'s home ground. <Is it, for example, that {sei gleki ko'a broda} means something like {ko'a broda ije mi gleki la'e di'u}? In that -- and any variant that fits what I would take "describe" to mean (something propositional at a min) -- then it always turns out to affect truth values and so is not much like a UI at all.> Officially (I notice belatedly) , it is {ko'a broda .i ko'a gleki}, two separate sentences, each with a truth value and each with {ko'a} as subject. This makes "that's not true" a bit hard to apply directly to the original sentence and also leaves open whether it is being broda that makes ko'a happy. So it is not exactly like a UI. I wonder about (to get back to the original issue) {sei cumki}. The structure of {cumki} makes the sentence (or the state it describes) the only plausible subject despite the pattern laid down. On the other hand, give the sentence is true, that it is possible is redundant. Still, if the sentence is false, that it is (or was) possible is useful additional information. I still think we need to think this through a bit more. |