From - Thu Feb 15 10:22:47 1996 Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id GAA21508 for ; Thu, 15 Feb 1996 06:48:36 -0500 Message-Id: <199602151148.GAA21508@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id 5CE433DE ; Thu, 15 Feb 1996 6:14:44 -0500 Date: Thu, 15 Feb 1996 13:13:23 +0200 Reply-To: veion@XIRON.PC.HELSINKI.FI Sender: Lojban list From: Veijo Vilva Subject: Re: sera'aku GEN: almost-PROPOSAL: intervals X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan X-Mozilla-Status: 0011 Content-Length: 2013 la lojbab cusku di'e > Date: Thu, 15 Feb 1996 05:43:53 -0500 > From: Logical Language Group > Subject: Re: sera'aku GEN: almost-PROPOSAL: intervals > But seeing this example does cause something to click in my mind, that > generalizes the problem. I do not yet have a solution. But what we seem > to be facing is a situation where we want a sumti tcita to attach to, not one, > but 2 sumti, each with a somewhat different relationship to the seubordinate > bridi implied by the tcita. We already HAVE a parallle problem that is > much clearer: what if you want to attach a 2nd sumti onto the implied > relationship of a BAI tcita. [...] > I would be interested in grammar and cmavo proposals that address this type of > multiple sumti linking, and I would rather see a new cmavo used for the link > than overloading an existing one, unless you can clearly establish that the > grammatical and semantic role is identical. I am NOT sure that relative clauses > are a model for this becuase they suggest that one of the sumti being attached > to the tag is subordinate to the other, which is not the general case. > > Veijo has done an admirable job of turning ideas into concrete YACC proposals, > even when he is not sure he favors the proposal. I commend these efforts, > and would like to know in particular what he can do with this idea. Thanks! Here is a very general solution which passes YACC and thus shows that almost anything goes because, if BO works as glue, then any odd new cmavo will work. modifier_82 : mod_head_490 gap_450 C | mod_head_490 linked_sumti_84 ; + linked_sumti_84 : sumti_90 + | linked_sumti_84 BO_508 sumti_90 + ; -- co'o mi'e veion --------------------------------- .i mi du la'o sy. Veijo Vilva sy. ---------------------------------