sarge@thirdi.UUCP (Sarge Gerbode) (08/12/87)
Awhile back, when we were discussing knowledge and belief, a couple of people, as I recall, suggested that knowledge was justified true belief. Since then, I have spoken to a philosopher friend, who pointed out that this criterion doesn't hold water, so I thought I'd share his thought with you. The first example is paraphrased from an article by Edmund C. Gettier called "Is justified true belief knowledge?": Consider the case where Smith has good reason for thinking that Jones owns a Ford. Imagine further that Smith has another friend, Brown, whose whereabouts he doesn't know. In this case, Smith has justified belief for: 1. Either Jones owns a Ford or Brown is in Boston. 2. Either Jones owns a Ford or Brown is in Barcelona 3. Either Jones owns a Ford or Brown is in Brest-Litovsk Each of these disjunctions is entailed by "Jones owns a Ford", for which Smith, *ex hypothesi* has justified belief. Now, suppose that Jones in fact doesn't own a Ford but is driving a rented car, and suppose further that Brown just happens to be in Barcelona. Smith then has justified true belief that "Either Jones owns a Ford or Brown is in Barcelona", but would we say that he *knows* this? A second counter-example is a causal one (not from the same article). Suppose a certain pyromaniac has a box of "Shur-Fire" matches and that he has very good reason for believing that if he strikes a match it will light. Say he strikes a match and it does, in fact, light, but it's actually a defective match and would not have lit except that a burst of "Q-radiation" came in from the stratosphere and hit it just as he scratched it across the box (or suppose an improbable concentration of high-energy molecules just happened to congregate in a certain location on the match, causing it to light). He had justified true belief that it would light, but did he *know* it would light? Each of these examples uses a chain of inferences (logical or causal) in which a term earlier than the final one in the chain is one for which a person has justified belief but which is not in fact true, and an ultimate term that happens to be true and deducible from, or could be caused by, the term that is false. In order to avoid this problem, it seems that one would have to add another proviso to the "justified true belief" criterion, namely that one would require justified true belief also in each of the non-ultimate terms in a causal or explanatory chain. A possible difficulty with this formulation is that the notion of where an explanatory chain or a causal chain *starts* is somewhat up in the air, to say the least. So you would have a lot of work showing that you had justified true belief for anything all the way back to the beginning of the chain. So you would have trouble, using this criterion, saying that you know anything. -- "Absolute knowledge means never having to change your mind." Sarge Gerbode Institute for Research in Metapsychology 950 Guinda St. Palo Alto, CA 94301 UUCP: pyramid!thirdi!sarge
jiml@alberta.UUCP (Jim Laycock) (08/13/87)
In article <90@thirdi.UUCP> sarge@thirdi.UUCP (Sarge Gerbode) writes: |Awhile back, when we were discussing knowledge and belief, a couple of people, |as I recall, suggested that knowledge was justified true belief. Since then, |I have spoken to a philosopher friend, who pointed out that this criterion |doesn't hold water, so I thought I'd share his thought with you. The first |example is paraphrased from an article by Edmund C. Gettier called "Is |justified true belief knowledge?": | [discussion of Gettier counter-example and pyromaniac causal example] |Each of these examples uses a chain of inferences (logical or causal) in which |a term earlier than the final one in the chain is one for which a person has |justified belief but which is not in fact true, and an ultimate term that |happens to be true and deducible from, or could be caused by, the term that is |false. |In order to avoid this problem, it seems that one would have to add another |proviso to the "justified true belief" criterion, namely that one would |require justified true belief also in each of the non-ultimate terms in a |causal or explanatory chain. A possible difficulty with this formulation is |that the notion of where an explanatory chain or a causal chain *starts* is |somewhat up in the air, to say the least. So you would have a lot of work |showing that you had justified true belief for anything all the way back to |the beginning of the chain. So you would have trouble, using this criterion, |saying that you know anything. |Sarge Gerbode This is just what I was saying some time ago. I'll accept the definition of knowledge as "justified true belief", but then I'll insist (for the very reasons you pointed out above) that, given this definition, knowledge is not attainable. But we shouldn't despair at this fact, nor should we lapse into relativism. There's alot we can do with well-substatiated and well-corroborated evidence. We must also be careful, in our discussion, to steer away from the colloquial definition of knowledge. Our `knowledge' is a technical term, and is to be divorced from "acquainted with" or "have a pretty good idea about" concepts. One thing that has bothered me about these discussions, however, is that people have been taking for granted the following two propositions: 1. Bel(p) & Bel(q) -> Bel(p & q) 2. Bel(p) or Bel(q) -> Bel(p or q) These seem to give rise to many of our so-called epistemic paradoxes, so that in itself might be reason to cast them in doubt. One would have a much weaker epistemic logic if one were to abandon these principles, but I don't thing they're psychologically appealing. Alot goes on in the `->' st -- Jim Laycock Philosophy grad, University of Alberta alberta!Jim_Laycock@UQV-MTS OR decvax!bellcore!ulysses!mhuxr!mhuxn!ihnp4!alberta!cavell!jiml
rjf@eagle.UUCP (08/15/87)
Summary: Expires: Sender: Followup-To: In article <90@thirdi.UUCP> sarge@thirdi.UUCP (Sarge Gerbode) writes: >[..about knowledge as justified true belief..] > >In order to avoid this problem, it seems that one would have to add another >proviso to the "justified true belief" criterion, namely that one would >require justified true belief also in each of the non-ultimate terms in a >causal or explanatory chain.. >..So you would have trouble, using this criterion, >saying that you know anything. In article <142@cavell.UUCP> jiml@cavell.UUCP (Jim Laycock) writes: > > This is just what I was saying some time ago. I'll accept the definition >of knowledge as "justified true belief", but then I'll insist (for the very >reasons you pointed out above) that, given this definition, knowledge is >not attainable. But we shouldn't despair at this fact, nor should we >lapse into relativism. There's alot we can do with well-substatiated >and well-corroborated evidence. > We must also be careful, in our discussion, to steer away from the >colloquial definition of knowledge. Our `knowledge' is a technical term, and >is to be divorced from "acquainted with" or "have a pretty good idea about" >concepts. >-- > Jim Laycock Philosophy grad, University of Alberta As the person who first (in this group :-) I think :-) ) suggested the justified true belief definition, I'm totally with Jim on this one. I long ago came to the conclusion that knowledge, in strict terms is impossible, though I used another route to get to this point. I won't go into that here - it would make for a rather long posting, and I'd probably get flamed to death from people who are tired of consciousness/subjectivity/ objectivity discussions in mod.ai or whatever it is now. Suffice it to say that I have reason to believe that objectivity, strictly speaking, is impossible, and that knowledge is an intrinsically objectivist (using that word very loosely) concept. Maybe it was a mistake to say even that much. Ah well.. I hereby give notice that I *WILL NOT* reply to articles or email on this subject, unless in the meantime I think of a very simple and unsubtle explanation of it :-) Unlikely, I'm afraid. If anyone is interested, hunt up `Mortal Questions' by Thomas Nagel, pub. Harvester Press (in UK), 1978 (I think), and read the last two chapters. If you're interested and open- minded I can promise an utterly fascinating read. >"Absolute knowledge means never having to change your mind." > >Sarge Gerbode Does anyone still believe in absolute knowledge? (Outside of religious experience, PLEASE) Robin rjf@ukc.ac.uk ..mcvax!ukc!rjf
sarge@thirdi.UUCP (Sarge Gerbode) (08/25/87)
Re: article <3259@eagle.ukc.ac.uk> rjf@ukc.ac.uk (Robin Faichney): I read with interest your discussion of philosophy, and I agree that your definition is a redefinition. I guess my concern was that there was some baby being thrown out with the bathwater. I tend to agree with *somebody* who said that reducing philosophy to a discussion of meanings was to fuse philosophy and semantics or philosophy and linguistics. Semantics and linguistics are, of course important, as is mathematics and logic, and my concept of philosophy was that it was a rational way of approaching the "Big Issues" of life, rather than relegating these to blind faith or authority. Things like what is right and wrong, what is the good, the bad, and the ugly? I am optimistic in that I think there *are* ways of approaching these issues rationally and finding answers that enhance the quality of life or act as a guide to life. I wouldn't be comfortable with a definition of philosophy that omitted these things. In fact, I *wasn't*. I left philosophy when I observed it was getting too tied up in linguistic analysis. If we accept your definition of philosophy, then we must find another term to describe the rational pursuit of wisdom. And I don't think that term is "psychology" or "art appreciation". Nor do I think it's "religion". >As the person who first (in this group :-) I think :-) ) suggested the >justified true belief definition, I'm totally with Jim on this one. Re: "justified true belief". So *you're* the culprit! >I long ago came to the conclusion that knowledge, in strict terms is >impossible, though I used another route to get to this point. I won't go >into that here - it would make for a rather long posting, and I'd probably >get flamed to death from people who are tired of consciousness/subjectivity/ >objectivity discussions in mod.ai or whatever it is now. Suffice it to say >that I have reason to believe that objectivity, strictly speaking, is >impossible, and that knowledge is an intrinsically objectivist (using that >word very loosely) concept. I agree with you about objective truth. It isn't just to be a devil's advocate that I've been trying to work out a *subjective* criterion. >Does anyone still believe in absolute knowledge? (Outside of religious >experience, PLEASE) It seemed to me that some people were talking as though they did. -- "Absolute knowledge means never having to change your mind." Sarge Gerbode Institute for Research in Metapsychology 950 Guinda St. Palo Alto, CA 94301 UUCP: pyramid!thirdi!sarge
sarge@thirdi.UUCP (Sarge Gerbode) (08/25/87)
In article <142@Cavell.UUCP> jim@cavell.UUCP (Jim Laycock) writes: >We must also be careful, in our discussion, to steer away from the colloquial >definition of knowledge. Our 'knowledge' is a technical term, and is to be >divorced from "acquainted with" or "have a pretty good idea about". If it's a technical term, then surely it should have a strict definition, and I haven't seen one being offered recently. Somebody should stick his neck out. I also don't see why "knowledge" in a scientific context is any more or less knowledge than the "knowledge" that we have in our everyday lives. I am in favor of defining things; I think the proferred definition should cover both scientific and non-scientific usages, or else we should use two different words or phrases to describe the different forms of knowledge and a strict criterion should be offered for separating them. I rather liked *my* definition (Why not? Everyone likes his own ideas!): knowledge is certainty. I realize this is a subjective definition, but our attempts to find an objective one have so far failed. I'm open to suggestions as to how to define "knowledge" objectively. -- "Absolute knowledge means never having to change your mind." Sarge Gerbode Institute for Research in Metapsychology 950 Guinda St. Palo Alto, CA 94301 UUCP: pyramid!thirdi!sarge
rsl@ihlpl.ATT.COM (Richard S. Latimer ) (08/28/87)
> From: Sarge Gerbode > I am in favor of defining things; ... > I rather liked *my* definition (Why not? Everyone likes his own ideas!): > knowledge is certainty. A proper definition needs to identify the `genus' or class to which the concept belongs and the `differentia' or differences which distinguish the concept from others in that class. "Knowledge is certainty" is thus NOT a proper definition (it only indicates how you feel with regard your `knowlegde notion'). How about?: genus = a hierarchial structure of inter-related thoughts and differentia = derived from actual sensations/perceptions (i.e. from reality) integrated via logical (non-contradictory) methods (i.e. through reasoning). Thus: definition = a hierarchial structure of inter-related thoughts, derived from actual sensations/perceptions integrated via logical methods. Or a `short form': "a mental grasping of facts of reality". How do you like that definition? If it is unsatisfactory, can you identify a clearer or more precise genus and/or differentia? -- Eudaemonia, Richard S. Latimer [(312)-416-7501, ihnp4!ihlpl!rsl]
sarge@thirdi.UUCP (08/30/87)
In article <2606@ihlpl.ATT.COM> rsl@ihlpl.ATT.COM (Richard S. Latimer) writes: >"Knowledge is certainty" is thus NOT a proper definition (it only >indicates how you feel with regard your `knowlegde notion'). I don't understand what you mean by "knowledge notion". >A proper definition needs to identify the `genus' or class to which >the concept belongs and the `differentia' or differences which >distinguish the concept from others in that class. > > >How about?: genus = a hierarchial structure of inter-related thoughts > and differentia = derived from actual sensations/perceptions > (i.e. from reality) integrated via logical > (non-contradictory) methods (i.e. through reasoning). >Thus: definition = a hierarchial structure of inter-related thoughts, > derived from actual sensations/perceptions > integrated via logical methods. >Or a `short form': "a mental grasping of facts of reality". > >How do you like that definition? If it is unsatisfactory, can you >identify a clearer or more precise genus and/or differentia? I can't say I really understand what you mean. What's "a mental grasping of facts of reality"? "Knowing things"? If this is the case, then I think that's quite a different thing from giving a definition. Some definitions might contain genera and differentia (like the classical "featherless biped"), but I don't think they *all* do. I prefer a simpler definition of "definition": Let's call an object that is intended to convey a concept (a word, phrase, symbol, picture, gesture, etc.) a "token". Then: Token A is a definition of token B iff Token A and token B refer to the same concept. Note that the relation "is a definition of" is commutative. So B would also be a definition of A. Latimer's notion of definition would not cover, for instance, ostensive definitions. In the exchange: Q: "What is 'Jane'?" A: "[Points to Jane].", it is hard to think of hierarchical structures of genera. Note that even this ostensive definition is commutative: Q: "[Points to Jane]?" A: "That is Jane." -- "Absolute knowledge means never having to change your mind." Sarge Gerbode Institute for Research in Metapsychology 950 Guinda St. Palo Alto, CA 94301 UUCP: pyramid!thirdi!sarge