Return-Path: <@FINHUTC.HUT.FI:LOJBAN@CUVMB.BITNET> Received: from FINHUTC.hut.fi by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0rCZte-00007FC; Tue, 29 Nov 94 23:15 EET Message-Id: Received: from FINHUTC.HUT.FI by FINHUTC.hut.fi (IBM VM SMTP V2R2) with BSMTP id 8337; Tue, 29 Nov 94 23:15:44 EET Received: from SEARN.SUNET.SE (NJE origin MAILER@SEARN) by FINHUTC.HUT.FI (LMail V1.1d/1.7f) with BSMTP id 8333; Tue, 29 Nov 1994 23:15:42 +0200 Received: from SEARN.SUNET.SE (NJE origin LISTSERV@SEARN) by SEARN.SUNET.SE (LMail V1.2a/1.8a) with BSMTP id 6772; Tue, 29 Nov 1994 22:12:05 +0100 Date: Tue, 29 Nov 1994 14:53:44 -0600 Reply-To: David Bowen Sender: Lojban list From: David Bowen Subject: Re: veridicality in grammar X-To: lojban@cuvmb.cc.columbia.edu To: Veijo Vilva In-Reply-To: Your message of "Mon, 28 Nov 1994 09:02:56 EST." <199411292018.OAA19779@timbuk.cray.com> Content-Length: 1929 Lines: 47 >I am using `grammatical' here in the sense that fluent speakers of a >language will tell you whether an utterance is grammatical. For >example, speakers of English will tell you that the following is >well formed: > > Green ideas sleep furiously. > >They consider the following to be ill formed: > > *Sleep green furiously ideas. > >Also, English speakers will tell you the following is ill formed: > > *Green ideas sleeps furiously. > >Veridicality is not a criterion for whether English utterances are >grammatical. The following is well formed, even if false in the >context of the current conversation: > > Tomorrow, Hannibal will cross the Alps. > >Lojban is different. The following is a grammatical use of {lo} if >and only if the cat seen is `for real' in the context of the current >conversation: > > .i la dgorj ca ca'a viska lo mlatu > >However, the utterance is not grammatical if the cat is not `for >real'. If the cat is not `for real', but is something you are >designating as a cat, then the grammatical categorizer is {le}. I'm afraid I must respectfully disagree here, though my disagreement may be a disagreement about the definition of well-formed. My definition is that "well-formed" is equivalent to "is syntactically correct". In other words, if I can build a parse tree for it using the current Lojban grammar it's well-formed. I would categorize errors in the usage of "lo" and "le" as semantic errors. ".i la dgorj ca ca'a viska lo mlatu" will parse whether of not "lo mlatu" really is a cat. But it will be meaningless if "lo mlatu" isn't a cat. Since veridicality isn't usually a concern in English it's difficult to construct an analog. References to "deciduous pines", "colorless green objects" or "sunsets in the eastern sky" would be possible examples of syntactically valid phrases that lack meaning. I'm sure others can construct other examples. David Bowen