From: Logical Language Group Message-Id: <199502241818.AA20161@access2.digex.net> Subject: Re: Carterian formula (was: Gricean formula?) Date: Fri, 24 Feb 1995 13:18:22 -0500 (EST) Content-Transfer-Encoding: 7bit X-From-Space-Date: Fri Feb 24 13:18:29 1995 X-From-Space-Address: lojbab > I don't understand why you think it's useful to neglect animals with a heart > .onai a kidney. My point is that such an animal is only marginally an animal: it tends to be dead, or on total life support. But the fact that {x | x has a heart} and {x | x has a kidney} are the same set, does not mean that having a heart and having a kidney should be reckoned identical properties. Quine uses this example to discredit the idea of properties, but I don't think that will work for natural-language use, where we think in terms of objects and their properties. > Can you come up with a "minimal pair" of sentences that > might exist in a language, differing only in their use of predicates meaning > "x1 has a heart" and "x1 has kidneys", *without* relying on dissection, > organ transplants, unusual species, etc? How about a pair of sentences > using two predicates whose referent sets are guaranteed to be the same? > (i.e. "x1 has skin", "x1 has a skin color") Sure, and so can you, but I'm not sure what point you're making. The point is that such pairs of sentences have the same truth conditions, but they don't mean the same thing, in any usual sense of "mean". > Agreed. But what *does* "lo ratcu" mean in Lojban, all by itself? It's an incomplete utterance, so it is either used to complete a previous utterance (typically, but not necessarily, a question), or it tends to provoke a completion (possibly a question) itself. Thus, if you come up to me and say "lo ratcu" I will probably respond "lo ratcu cu ?mo". -- John Cowan sharing account for now e'osai ko sarji la lojban.