From thinkit8@lycos.com Mon Dec 24 16:54:20 2001 Return-Path: X-Sender: thinkit8@lycos.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_0_1_3); 25 Dec 2001 00:54:20 -0000 Received: (qmail 86913 invoked from network); 25 Dec 2001 00:54:19 -0000 Received: from unknown (216.115.97.172) by m2.grp.snv.yahoo.com with QMQP; 25 Dec 2001 00:54:19 -0000 Received: from unknown (HELO n18.groups.yahoo.com) (216.115.96.68) by mta2.grp.snv.yahoo.com with SMTP; 25 Dec 2001 00:54:19 -0000 Received: from [216.115.96.38] by n18.groups.yahoo.com with NNFMP; 25 Dec 2001 00:54:28 -0000 Date: Tue, 25 Dec 2001 00:54:17 -0000 To: lojban@yahoogroups.com Subject: Re: Binary Language Message-ID: In-Reply-To: <154.64907c1.29589e49@aol.com> User-Agent: eGroups-EW/0.82 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Length: 3154 X-Mailer: Yahoo Groups Message Poster From: "thinkit41" X-Originating-IP: 12.224.27.33 X-Yahoo-Group-Post: member; u=71054096 X-Yahoo-Profile: thinkit41 X-Yahoo-Message-Num: 12697 --- In lojban@y..., pycyn@a... wrote: > In a message dated 12/24/2001 1:36:10 AM Central Standard Time, > rob@t... writes: > > > > How about concepts like "between" or > > "combine" where the x2 and x3 are interchangeable? Or would you simply > > leave those out of your language? > > > > Well, thinkit would say that "between" (and probably "combine") is easy. "a > is between b and c" is just "a is to the side of b with a to the side of c" > and if you said that this would allow all sorts of arrangements where a is > NOT between b and c, he would claim that those all required some other > condition but that "between" was the natural reading of the simplest one. If > you strike that down, then he may eventually be driven to the place binarists > usually end up, making arguments of pairs of objects -- ordered or unordered > as the need may be (and pairs of pairs or of a pair and an object, and so on, > as needed). As I said, "between" would be a one argument verb. Try A between (tag) B has location (tag) C has location. Since tags aren't ordered anyway, this makes sense. > thinkit: > is, man attempts to (sentence) man give book to cat (end sentence) > (tag) thwarted by (sentence) dog take book from man. The sentence I > gave could have no other reasonable meaning than the man giving a > book to a dog (assuming no other tags are given). The second > mention of the book (in the tag) isn't even necessary> > > Well, but the first solution requires a way to make sentences into arguments > -- which ought to be interesting in this system -- and then a way of saying > that the man gave the book to the cat, which is just the original problem > again, not solved by saying that it is part of the solution to some other > problem (ditto that the dog takes the book from the man). I think that > dropping the second reference to the book requires a convention that one > probably does not want to use: that the missing argument is the nearest > extrasentential reference or the one in the same place or some such. The > exceptions will be more numerous than the cases and the possibility for error > rather large (e.g., taking the gap to mean there is nothing there - - a very > real case -- or that it is an unspecified something -- also a real case). > since there has to be something in the 2nd argument place, it might as well > be what is intended. If there is a reasonable way you could be misunderstood, the option is always there to use a reference. > identified as objects. > > > But it sure seems to be two sentences both of which (and more beside, I > think) have to be true for the intended meaning. If either part fails then > the whole fails: a clear case of conjunction, not subordination. > HOW are the man, dog and book identified as arguments rather than predicates? > (Lord, your terminology is getting even me talking nonsense). It's one sentence that has a sentence contained in it.