From araizen@newmail.net Sat Oct 21 13:22:45 2000 Return-Path: X-Sender: araizen@newmail.net X-Apparently-To: lojban@egroups.com Received: (EGP: mail-6_1_0); 21 Oct 2000 20:22:45 -0000 Received: (qmail 21574 invoked from network); 21 Oct 2000 20:22:45 -0000 Received: from unknown (10.1.10.27) by m8.onelist.org with QMQP; 21 Oct 2000 20:22:45 -0000 Received: from unknown (HELO out.newmail.net) (212.150.51.26) by mta2 with SMTP; 21 Oct 2000 20:22:44 -0000 Received: from default ([62.0.180.47]) by out.newmail.net ; Sat, 21 Oct 2000 22:24:09 +00:00 To: lojban@egroups.com Date: Sat, 21 Oct 2000 22:22:19 +0200 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: RE:literalism Reply-to: araizen@newmail.net Priority: normal In-reply-to: <972131813.25843@egroups.com> X-mailer: Pegasus Mail for Win32 (v3.11) Message-ID: <97219225001@out.newmail.net> From: "Adam Raizen" la pycyn cusku di'e > Even the literalist > allows that there are maybe a dozen (or maybe fewer, but more > than one) rules and NO flag for which rule is used. Actually, I've always thought it would be useful if we could optionally indicate at least when the tanru/lujvo is asymmetrical, for the situation when a word which looks like an adjective (i.e. symmetrical) is used asymetrically. The symmetrical version of a tanru/lujvo can be expressed with 'je' or 'joi', and I suppose that that leaves the asymmetrical version as the default (which is considered the more 'justified' version according to Nick Nicholas' paper anyway.) > Yet somehow > some rules are canonized and others are suspect, on no stated nor > defended basis: this is better (though ugly and hard to deal with) > that that (though immediately clear and memorable) because it uses > my favorite rule. Phooey! I'd love more rules, but they have to deal with the place stucture. They have to be formalizable. (I suppose I'm being a literalist.) Any rule which relies on omitted words or figures of speech ("metaphors") is hard to understand and probably culturally and idiolectally dependent. For example, a pattern that I've sometimes seen is when the seltau modifies the entire bridi, as in e.g. 'spaji nerkla' (le nu nerkla cu spaji). Would it work to consider the place stucture of 'pajykemnerkla' as 'n1 n2 n3 n4 n5 s2 (n=nerkla)'? Would it work to define this as type of tanru as broda zei brode = brode1 brode2 brode3 brode4 brode5 broda2 broda3 broda4 broda5? Or maybe there's another way to analyze it. co'o mi'e adam