Date: Thu, 1 Jan 1998 21:22:04 -0500 (EST) Message-Id: <199801020222.VAA17015@locke.ccil.org> Reply-To: Steven Belknap Sender: Lojban list From: Steven Belknap Subject: X-To: LOJBAN@cuvmb.cc.columbia.edu To: John Cowan X-Mozilla-Status: 0001 Status: RO Content-Length: 4855 Lines: 92 Ashley: >>It seems to me that one must either fuzzify with >>, specify the epistemology or method by which certain knowledge is >>claimed, or use a bridi which reports "Just the facts, M'am," such as: >... >Oh, so "djuno" implies '_certain_ knowledge' now? Lets just stick to the example, or we will get off into the deep mud. "djuno" refers to "known facts". Of the 20,000+ attending a Cubs game, some will see the play differently than others. But the only views that matter (according to the rules of the game) is that of the umpires, particularly the home plate umpire. A perfectly valid use of would be: because the umpire's knowledge is *defined* as being peremptory (and the rules of baseball are the implicit epistemology). There are other formal systems which would also be appropriate for application of . These are such things as mathematics, law, religion, codified social protocols and many others. The usage I am proposing for is really not very restrictive at all, and usefully distinguishes between formal systems knowledge and other types of ideas. In the stadium where Mark sits, there may be considerable disagreement about whether or not the pitch was a strike or a ball or was caught. There is no formal mechanism for deciding the matter among these observers, (although fisticuffs seems to be popular as an informal mechanism), so it is hard for me to see how djuno would be used. Of course, one could *define* a craft or method for , and there is a sumti for that built in to . The X3 place is to be filled by a meme which defines a decision function. >Bear in mind that the >range of things one might have non-fuzzily justifiably certain knowledge >of is extremely limited, and wouldn't include your 'possibly true' example > > "Steven knows that Jorge asserts that Lojbab goes to the store." There is great irony in your statement, although I don't know if you were reading the list during the Great Fuzzy Debates of last year. I certainly agree strongly with you that fuzzy statements deserve a very prominent place in lojban, as predicate logic differs substantially from natlang meaning, and fuzziness is a better reflection of the meaning of most natlang utterances. I argued vociferously for inclusion of syntax which facilitated fuzzy statements in the grammer of lojban. Much of it was there, (although perhaps unrecognized), the rest can probably be handled by , which is in the grammer. (Hooray!) I suggest that the terrible flaw in Samuel Delaney's Babel-17 was that there was no fuzzifier, and that the big improvement in Babel-18 was that they added . :-) > >since Steven can never rule out having misheard, etc. Your definition of >"djuno" doesn't seem particularly useful, and I would suggest that one >matching a more usual English definition of 'know' (as indeed the >gismu-list suggests) might be better. I am trying very hard to match the definition as given in the very short dictionary entry. However, it is clear that only the tip of the iceberg is covered by the entry for : djuno [ jun ju'o ] know x1 knows fact(s) x2 (du'u) about subject x3 by epistemology x4 [words usable for epistemology typically have a du'u place; know how to - implying knowledge of method but not necessarily having the ability to practice (= tadjyju'o)] (cf. know/familiar with: se slabu, na'e cnino, na'e fange; cmavo list du'o, cilre, certu, facki, jijnu, jimpe, senpi, smadi, kakne, birti, mipri, morji, saske, viska) > >In any case, I think there are very few fields of discourse in which one >can assume that every assertion will be perfectly true or false >(mathematical proof is one, no others come to mind). Everything from Baseball to Monopoly to Algebraic topology, from Islam to Confucianism has a codicile which defines the body of a particular epistemology. Even codified science would qualify, I suppose, although in my view codified science is pseudoscience. That seems like a rather large subset of the universe of discourse to me! Wouldn't cricket qualify as well? >Everywhere else, one >has to allow for fuzziness, so it's nothing special for assertions >involving "djuno" to be fuzzy any more than those involving "crino". Agreed. That was why I briefly mentioned the fuzzy issue before dissecting the other issues related to . Perhaps the standard context in lojbanistan will be fuzzy, and the statements made in lojban will be shorthand for fuzzified statements. I don't know that this is desirable, and I would prefer to explicitly fuzzify until it is clear to me that your assertion is true. co'omi'e la stivn Steven Belknap, M.D. Assistant Professor of Clinical Pharmacology and Medicine University of Illinois College of Medicine at Peoria