From sentto-44114-3249-962046444-mark=kli.org@returns.onelist.com Mon Jun 26 19:06:21 2000 Return-Path: Delivered-To: shoulson-kli@meson.org Received: (qmail 1441 invoked from network); 26 Jun 2000 19:06:20 -0000 Received: from zash.lupine.org (205.186.156.18) by pi.meson.org with SMTP; 26 Jun 2000 19:06:20 -0000 Received: (qmail 11726 invoked by uid 40001); 26 Jun 2000 19:07:26 -0000 Delivered-To: kli-mark@kli.org Received: (qmail 11723 invoked from network); 26 Jun 2000 19:07:25 -0000 Received: from hh.egroups.com (208.50.144.88) by zash.lupine.org with SMTP; 26 Jun 2000 19:07:25 -0000 X-eGroups-Return: sentto-44114-3249-962046444-mark=kli.org@returns.onelist.com Received: from [10.1.10.35] by hh.egroups.com with NNFMP; 26 Jun 2000 19:07:23 -0000 Received: (qmail 23466 invoked from network); 26 Jun 2000 19:07:23 -0000 Received: from unknown (10.1.10.27) by m1.onelist.org with QMQP; 26 Jun 2000 19:07:23 -0000 Received: from unknown (HELO tele-post-20.mail.demon.net) (194.217.242.20) by mta2 with SMTP; 26 Jun 2000 19:07:22 -0000 Received: from kindness.demon.co.uk ([158.152.216.198] helo=arac) by tele-post-20.mail.demon.net with smtp (Exim 2.12 #2) id 136eE2-000Bbm-0K for lojban@egroups.com; Mon, 26 Jun 2000 19:07:18 +0000 To: "Lojban List" Message-ID: X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Importance: Normal From: "Colin Fine" MIME-Version: 1.0 Mailing-List: list lojban@egroups.com; contact lojban-owner@egroups.com Delivered-To: mailing list lojban@egroups.com Precedence: bulk List-Unsubscribe: Date: Mon, 26 Jun 2000 20:13:45 +0100 Subject: [lojban] Complements and adjuncts Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit The botpi question raises in my mind a more general question about Lojban and grammatical theory: I'm sure it must have been raised before, but I don't recall it. The question is, does Lojban distinguish complements and adjuncts? At first sight, the answer is yes: by design, the defined terbri of a selbri are complements (i.e. its meaning is incomplete without them), and any additional terbri attached by sumtcita must be adjuncts (optional extras specification, but not an essential part of the meaning). There are two problems with this analysis. The minor one is the idea I am sure I have seen stated (though I cannot find it in TCLL) that there is no essential difference between the defined terbri and additional ones - it's just a syntactic convenience that (for zipfean reasons I suppose) you can use this predefined set of arguments without tagging them. The other is that one of the reasons why the distinction between complements and adjuncts is significant is that in languages generally (I hesitate to say 'all languages') complements cannot come between the head and its adjuncts. (You can't say "the bottle with handles of water"). But Lojban lets us permute all the terbri with gay abandon. Of course, there is no law that says that Lojban has to obey any rules at all from natural language. But we should at least be aware if we are doing so - if only to explore how comprehensible we are when we do so. **************************************************************************** **** Colin Fine - Paradise Green Promotions - 01274 592696 Greyfriars Kirk House & Augustine's colin@paradise-green.co.uk **************************************************************************** **** -----Original Message----- From: pycyn@aol.com [mailto:pycyn@aol.com] Sent: 26 June 2000 02:19 To: lojban@egroups.com Subject: Re: [lojban] RECORD: containers As I said in the interest of NOT getting involved in this, lb has no better solution than English (in fact it doesn't have as good a one right now, but I don't want to say that either). But we are snagged on the horns of a dilemma, one which happily occurs (as of right now -- but there are other cases lurking) only when we mention that the bottle contains nothing. By the rules {ta botpi no da} is equipollent to {ta na botpi da} and thus follows from {ta tanxe}, on the plausible assumption that no box is a bottle. We can save the case by saying that the the equipollence does not hold, that at best we have an implication, but then we have (as xorxes notes) the page being a passer even though there is no one that he passes. The latter is as clearly wrong (or more so) as that {ta botpi noda} entails {ta botpi} is right, but they have the same logical form (and even the same English form, if you want to emulate the lb situation more closely). English avoids the problem by not having inherent places (as xorxes notes) and by putting the information in subordinate forms ("containing" or "of" or...) which can be factored out in logical expansions (but the problem could probably be recreated easily in cases where English did not have this device). For now we are stuck with the reasonable appeal to common sense to sort the cases out -- not a good position for a logic, though a common one for langauges (Robin the Turk had a note from someone who said logic was incompatible with language and it is cases like this that give the person's claim some force -- until they are taken care of.) This is not giving up on predicate logic, it is just not knowing how to make use of it correctly at this point. ------------------------------------------------------------------------ Get a NextCard Visa, in 30 seconds! 1. Fill in the brief application 2. Receive approval decision within 30 seconds 3. Get rates as low as 2.9% Intro or 9.9% Fixed APR http://click.egroups.com/1/5197/4/_/17627/_/961982372/ ------------------------------------------------------------------------ To unsubscribe, send mail to lojban-unsubscribe@onelist.com ------------------------------------------------------------------------ Accurate impartial advice on everything from laptops to table saws. http://click.egroups.com/1/4634/4/_/17627/_/962046444/ ------------------------------------------------------------------------ To unsubscribe, send mail to lojban-unsubscribe@onelist.com