Date: Fri, 9 Jan 1998 17:40:10 -0500 (EST) Message-Id: <199801092240.RAA26145@locke.ccil.org> Reply-To: "=?iso-8859-1?Q?Jorge_J._Llamb=EDas?=" Sender: Lojban list From: "=?iso-8859-1?Q?Jorge_J._Llamb=EDas?=" Subject: Re: knowledge and belief X-To: lojban To: John Cowan X-UIDL: 437f854a925e8d102a9cfee4ecbb735b Status: OR X-Mozilla-Status: 0011 Content-Length: 3583 X-From-Space-Date: Mon Jan 12 15:47:22 1998 X-From-Space-Address: - la markl cusku di'e >A pragmatically minded person has no problem saying >something like, "I don't know what her name is, but Genaro >does; go ask him." Neither should a logically minded person. What's illogical in that? And there isn't necessarily any incompatibility between pragmatism and logic. >This is outrageous to the logically minded person. "How >can you say that Genaro 'knows' her name when you haven't >even verified that what Genaro claims to 'know' is really >her name?" There could be many reasons. She might have told me that she told Genaro her name, and I may accept her word as true. What's illogical about that? >The logically minded person can cite the definition of >Lojban {djuno} - or English "know" - until the cows come >home, & the pragmatically minded person won't care, altho Is it really definitions that we're arguing? Do you really think that a pragmatically minded person will use "know" in English for something that is not true? All the examples of such use seem strained. Would anyone really say in normal English: "Santa doesn't exist, but my son knows that he does exist." That's just not how "know" is used in my experience. If you change "knows" to "is convinced" then the sentence becomes perfectly natural. >s/he may pretend to care for the sake of politeness. To >the pragmatically minded person, what matters is not how >the word is defined but, rather, what the speaker or >author meant when using it. Yes, and that's the best thing to do when we are using the language, in my opinion. I doubt that anyone taking part in this discussion would have a problem understanding the actual uses of {djuno} in actual Lojban texts. But I haven't seen any examples of actual Lojban text where "is convinced" would be a better translation than "knows". >The pragmatically minded person can refine the definition >with alternative glosses, trying to bring it closer to >hi/r intuitions about how authors & speakers really use >the word in practice, Who is the pragmatic side in this discussion? Lojbab wants to change (refine if you want) the definition to something that is not closer to the actual usage. Actual usage of {djuno} has been like English "know", not like English "is convinced". And there's a good reason for that: that's what the gismu list says that it means. >I could propose a compromise: to claim that someone else >is convinced of something, one need not consider it to be >true; Does anyone doubt that? I would use {birti} in Lojban to translate that. >to claim that someone else knows something, one must >consider its truth to be extremely likely, or at least >overwhelmingly plausible. What's the point? If you claim that something is blue, do you mean to make an absolute undisputable claim, or simply one that is extremely likely or overwhelmingly plausible? Of course there will always be margin for error, that doesn't matter. If you claim {djuno} then you presuppose {jetnu}. To what degree of certainty or likeness is a different question, perhaps to be dealt with by fuzzy logic. >But such compromises are, by >their very nature, acceptable only to the pragmatically >minded people who do not need them. Such compromises >might offer some illumination to the logically minded >people, but those are the very people who cannot allow >themselves to compromise. I'm not sure I follow. What would be the unyielding logically minded position here? It is not logic that requires {djuno} to mean "knows", it is simply what it has meant up to now, both in usage and in the definition. co'o mi'e xorxes