From a.rosta@lycos.co.uk Sat Sep 21 06:47:26 2002 Return-Path: X-Sender: a.rosta@lycos.co.uk X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_1_1_3); 21 Sep 2002 13:47:26 -0000 Received: (qmail 36671 invoked from network); 21 Sep 2002 13:47:25 -0000 Received: from unknown (66.218.66.217) by m7.grp.scd.yahoo.com with QMQP; 21 Sep 2002 13:47:25 -0000 Received: from unknown (HELO mailbox-15.st1.spray.net) (212.78.202.115) by mta2.grp.scd.yahoo.com with SMTP; 21 Sep 2002 13:47:25 -0000 Received: from oemcomputer (host213-121-71-121.surfport24.v21.co.uk [213.121.71.121]) by mailbox-15.st1.spray.net (Postfix) with SMTP id 407EE2088E for ; Sat, 21 Sep 2002 15:47:19 +0200 (DST) To: Subject: RE: [lojban] tu'o usage Date: Sat, 21 Sep 2002 14:48:59 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) Importance: Normal In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 From: "And Rosta" X-Yahoo-Group-Post: member; u=122260811 X-Yahoo-Profile: andjamin pc: > a.rosta@lycos.co.uk writes: > > << > > I don't see a difference between {pa lo su'o} and {pa lo ro}. What > am I missing? > >> > > The {ro}-{su'o} distinction goes back to a time when someone thought > that {ro}, "every," permitted the case of 0 of the whatsis and {su'o} > did not. The first part of this turned out to be false in the > official line (as in Logic), so there is not distinction and we > cannot meaningfully say {lo broda} if there are no broda, nor {lo no > broda} neither. I don't recall that being the official line -- indeed, according to what I do recall, the official line is what you say it isn't. Is it in Woldy somewhere? Anyway, even with nonimporting ro, I don't see how {pa lo su'o} differs from {pa lo ro}. > << > First off, let me note that {lo'e} serves as an adequate alternative > to {tu'o}. So I will recapitulate the reasons for preferring {lo'e} > or {tu'o} to {lo pa}. > >> > The Lojban {lo'e} might, but in a very twisted way -- the typical > member of a class of one is that one member, I suppose (but I bet I > could make a case for otherwise without doing much damage). On the > other hand, xorxes' {lo'e} (which is now yours as well, you say) I think it is inaccurate to speak of "the Lojban {lo'e}" in distinction to xorxes's and mine. It is not perverse to construe the ma'oste's gloss of {lo'e} as a clumsy attempt to capture the notion of generic reference, and what xorxes and I have been doing is trying to get a handle on generic reference. > << > 1. {lo pa} is sensitive to negation: whereas {tu'o broda na brode} > is unproblematic, it corresponds to {lo pa broda na ku brode}, not > to {lo pa broda na brode}. In my view, something that is sensitive > to scope adds complexity to the mental processing of the sentence. > >> > Actually, CLL never mentions this question in dealing with > quantifiers and negation. to be sure, sentences that have the size > of the set wrong are called false, but there is also no evidence I > could find that that would make the {na} denial true. I think it > wore likely that internal quantifiers are ... (I forget the > technical term, "filter?" probably not), that is, they are > preconditions that must be met for the sentences involving them to be > true (I think any sentencewhere this condition is not meant, even the > denial of one false for this reason, is false). Lojban has a > negation for that situation, {na'i}. So, {lo pa} is likely > impervious to {naku} movement, in a way that {pa lo}, for example, is > not (compare the case of {lo no} above, though this could just be a > problem of internal contradiction: "one or more out of none"). You're right that it has not been established whether the inner quantifier has the status of presupposition/conventional implicature -- i.e. being outside what is being asserted. However, since Lojban generally does not (or never, even?) use presupposition/conventional implicature, the default should be that the inner cardinality is being asserted. That doesn't stop anyone adducing arguments as to why this default should be overridden, though. > << > 2. {lo pa} makes a claim. I do not wish it to have to be the case > that whenever I talk about a du'u I also claim that there is only > one du'u. If I say {lo pa broda cu brode} I am claiming that > (i) something is broda and brode, and (ii) the cardinality of > lo'i broda is 1. But I want to be able to claim only (i). > >> > What is the fate of {tu'o broda} if there are moe than one broda? > Will every sentence containing the expression be false or only those > outside the scope of a {naku}? If the former, then it is exactly on > a par with {lo pa}. If the latter, then IT is the one making an > additional claim. If there are more than one broda then {tu'o broda} is ambiguous -- it is underspecified, and to form an interpretation the hearer will have to insert a quantifier. The same goes for when there is only one broda. In other words, {tu'o broda} is neither true nor false, because it expresses an incomplete logical formula. > << > 3. As I have already shown, the point of marking a singleton > category as a singleton category is to help the speaker and > hearer by signalling the greater logical simplicity. It runs > contrary to general principles of form--function iconicity to > signal simplicity of meaning by adding an extra meaningful word > (pa). > >> > But using a meaningless one (and so strictly dispensible) is OK? Yes. It is indispensible because the syntax requires a gadri or quantifier to be present at the start of a sumti. Ideally it would be possible to omit tu'o, but the syntax won't allow it; it's very much analogous to the use of dummy _there_ and _it_ in English to fill obligatory subject positions. --And.