From LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU Sat Mar 6 22:51:59 2010 Received: from YALEVM.YCC.YALE.EDU by MINERVA.CIS.YALE.EDU via SMTP; Tue, 25 May 1993 01:10:12 -0400 Received: from CUVMB.CC.COLUMBIA.EDU by YaleVM.YCC.Yale.Edu (IBM VM SMTP V2R2) with BSMTP id 0550; Tue, 25 May 93 01:09:23 EDT Received: from CUVMB.COLUMBIA.EDU by CUVMB.CC.COLUMBIA.EDU (Mailer R2.07) with BSMTP id 7382; Tue, 25 May 93 01:10:38 EDT Date: Tue, 25 May 1993 01:08:55 EDT Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: Re: Cowan on morphology X-To: conlang@diku.dk X-Cc: lojban@cuvmb.cc.columbia.edu To: Erik Rauch Status: O X-Status: X-From-Space-Date: Mon May 24 21:08:55 1993 X-From-Space-Address: @YaleVM.YCC.YALE.EDU:LOJBAN@CUVMB.BITNET Message-ID: I disagree with John that Lojban's word-making system makes ad hoc lujvo making difficult, based on first person experience. Even though I never memorized more than 50% of the rafsi (and I'm the only one I know who made enough of a systematic effort to even give a percentage), I was able to make words on the fly with no trouble, and for the most part, others were limited in understanding only when I used rafsi that they didn't know, and they learned the ones that I used most often rather quickly through ad hoc analysis in conversation (I am speaking of Sylvia and Athelstan especially in this regard, since neither had finished even mastering the gismu list at the time I was in my heyday of lujvo-making in converswation). Similarly, the people who have been producing Lojban text in quantity - especially Nick - seem to have had no inhibitiions abobout making and using lujvo. Where he din;t know, he sometimes guessed, and sometimes was even caught in his guesses, but many of the people who reviewed Nick's writing were able to understand and translate his results even though there were some rafsi errors and invalid forms, which shows that even for new people that HAVEN'T memorized the word lists, they can take apart lujvo on the fly, and presumably use context and guessing on the source gismu to interpolate meaning (I admit that I am worse at this than most - having formally memorized some of the rafsi, I am less tolerant of errors in rafsi because I reconstruct the word based on my knowledghe rather than guessing and hence get the wrong answer). Dozens of lujvo made it through the text reviews with errors, only to be caught mor erecently when Nora's glosser allowed us to be able to analyze lujvo in text for validity of form and to check to see if expansions were what the speaker intended. IN short, what I am saying is that the language WORKS in practice, with not very well-studied speakers, writers, and listeners/ readers. Hence arguments that the morphology is 'too complex' for ad hoc use are demonstrably false. ni'o Nora reminded me this evening, when I first mentioned this thread to her, that at one point she facetiously suggested a method of achieving perfect separation of words with no complex rules - just end all words with the same vowel, one that does not occur in the middles of words. This came up when we considered briefly changing the Lojban morphology as part of the spl split from JCB. Suppose, for example, that the final vowel of a Lojban word, whatver it was, was changed to the reduced (always unstressed) schwa, and some other means was found for hyphenating - probably the syllavbic r/n/l that is used for le'avla and some lujvo already. Yeah, the language would be simpler, and the rules would be too, but I don't think it would still be Loglan - too much of a different feel. And we still would have had to come up with rules for lujvo (even if they would have been simpler) le'avla (likewise), and cmavo (which would have a real problem if they also had to end in a schwa). In effect, of course, this silly idea of Nora's DOES occur in Loglan/Lojban - in names, where as a result there is almost no rules for internal structure (but of course no 2nd level of segregation, or even a way to tell a word turned into a name from a Lojban root as opposed to Lojbanized from another language). Lojban names alll end in a way that makes them distinct - consonant followed by pause. lojbab