From jjllambias@hotmail.com Fri Jul 06 18:17:34 2001 Return-Path: X-Sender: jjllambias@hotmail.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_2_0); 7 Jul 2001 01:17:33 -0000 Received: (qmail 2586 invoked from network); 7 Jul 2001 01:17:33 -0000 Received: from unknown (10.1.10.27) by l9.egroups.com with QMQP; 7 Jul 2001 01:17:33 -0000 Received: from unknown (HELO hotmail.com) (216.33.241.131) by mta2 with SMTP; 7 Jul 2001 01:17:32 -0000 Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Fri, 6 Jul 2001 18:17:32 -0700 Received: from 200.41.247.41 by lw8fd.law8.hotmail.msn.com with HTTP; Sat, 07 Jul 2001 01:17:32 GMT X-Originating-IP: [200.41.247.41] To: lojban@yahoogroups.com Bcc: Subject: Re: [lojban] Uses of Language Date: Sat, 07 Jul 2001 01:17:32 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: X-OriginalArrivalTime: 07 Jul 2001 01:17:32.0457 (UTC) FILETIME=[98F19990:01C10682] From: "Jorge Llambias" X-Yahoo-Message-Num: 8435 la pycyn cusku di'e >Nice! But you have about exhausted the forms: there remain only fragments >and ejaculations and observatives that I can think of -- and even >imperatives >are not a different *form*; the role is determined by the occurrence of >{ko}. In that sense, questions are not a different form either, since {ma}, like {ko}, is just another KOhA, and similarly for all the other question words. I was not talking about syntactic forms, classifying those is trivial in Lojban from the very definition of the language. I'm rather thinking of a semantic classification of the structures (which syntactically are all alike) but independent or prior to the use of those structures. Questions seem to be the easiest to identify (and maybe should be subdivided into fill-in-the-blank questions and true/false questions). The program would be to classify sentences of the form in terms of what the bridi refers to depending on the UI. (Then we can consider what happens with subclauses and so on, but we should start with the simplest cases.) I suppose I'm restricting myself to uses of bridi instead of considering all uses of language. I can think of at least 5 classes: 1- Assertions (probably most UI leave the bridi as an assertion, assuming that is its basic function). In these the bridi is used to refer to a situation which the speaker claims to hold in the world. (This includes fictional worlds, which I don't think have much to do with the possible worlds formulation. In fiction we talk as if the fictional world was real. It is fiction because we use the assertion forms.) 2- Questions. Here the bridi does not refer to one situation, but rather to a family of situations (sometimes with only two members) and the speaker wants to know which member of the family holds in the world. 3- Potentials. In these the bridi refers to a situation which may eventually hold in the world, but the speaker indicates no knowledge or belief that it does, only that it is compatible (ko, a'o, ai, e'o, la'a). 4- Counterfactuals. The bridi refers to a situation which the speaker to some degree indicates that does not hold in the world, I'm not sure to what degree (au, ba'u, da'i, je'unai, ju'onai). 5- Nonsensicals. The speaker indicates that the bridi does not refer to any situation (ki'a, na'i). I'm not sure whether in this scheme performatives should be in a different class than assertions. In them the bridi refers to a situation that comes to hold in the world as a result of or in conjunction with the utterance, so in a sense they could be considered to refer to a situation that holds in the world. I can't think of any other class at the moment, in terms of the holding status of the situations referred to by the bridi. mu'o mi'e xorxes _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.