ISSSSM%NUSVM.BITNET@CUNYVM.CUNY.EDU (Stephen Smoliar) (06/13/91)
In article <1991Jun12.221121.15828@watdragon.waterloo.edu> cpshelley@violet.waterloo.edu (cameron shelley) writes: >In article <1991Jun12.130817.3621@kingston.ac.uk> is_s425@kingston.ac.uk >(Hutchison C S) writes: > >>It seems to me that talk of 'partial truths', 'negotiation', and so on, may >>not get us very far. If I'm negotiating with you, I'm really just trying to >>tell you why you are (mostly) wrong and I am (mostly) right. If I adduce >>evidence to support my claims, then we may end up negotiating what counts as >>evidence. We're stuck in a hopeless regress. (Try telling one billion >>Christians or one billion Muslims they're wrong -- especially if it is >>perfectly obvious to you that Humanistic Buddhism is the only right way. Try >>negotiating with the Jehovah's Witness on your doorstep. Try telling the >>free market liberal about the unspeakable suffering and brutality that >>capitalism has wrought upon the cheap labour markets of the Third World.) > >What you seem to be saying is that the *process* of understanding (or >failing to understand) is very hard in difficult cases. Actually, I think that Chris may be saying more than that. I think he also seems to believe that there is some single, fixed resolution to negotiation. My own position is that this is a serious mistake. Any view of intelligence which does not take into account the ongoing nature of behavior is bound to miss the mark; and this includes the fact that negotiations are rarely (if ever) resolved absolutely. > >>To get things in context, despite the political flavour that my question may >>appear to have taken on, my main concern is with automatic knowledge >>acquisition >>from text (whatever kind of text it may be). My problem is: is knowledge >>representation going to be about an intelligent agent's models of the >>physical world or of speakers' reports about the world? This is a technical >>rather than a philosophical issue since it impinges directly on what kinds >>of inference and what sources of knowledge are relevant to the reasoning >>process. > >Like Carbonell's (and Hovy's) systems, a model of the physical world will >require 'objective' input at some point. Since this is not really possible, >I would select option b) you give above. The other possibility is to try to get the objectivity out of the first option. As I pointed out in my previous article, the REAL problem with those African headline lies in trying to assume a "generic" reader. Even Carbonell's liberals and conservatives are basically "generic." When they read about military action in Bhutan, they do not have to worry about running a business there. If we try to build intelligence agents which are less abstract and more subjective, we might be able to make some progress on Chris' first option. =============================================================================== Stephen W. Smoliar Institute of Systems Science National University of Singapore Heng Mui Keng Terrace, Kent Ridge SINGAPORE 0511 BITNET: ISSSSM@NUSVM "He was of Lord Essex's opinion, 'rather to go an hundred miles to speak with one wise man, than five miles to see a fair town.'"--Boswell on Johnson
ISSSSM@NUSVM.BITNET (Stephen Smoliar) (06/15/91)
In article <1991Jun14.111857.7374@kingston.ac.uk> is_s425@kingston.ac.uk (Hutchison C S) writes (after attributing my latest round of comments to Cam Shelley): > >Again, I wholly agree with you that you "are not talking about text-based >propositions which are true or false". I think you are talking about >something like 'perlocutionary effects'. Actually, I have tried to avoid the primitive practice of taming my demons by calling them by name! :-) > Of course the propositions expressed >by sentences will have effects on the behaviour (cognitive or physical) of >hearer-readers. That is merely a matter of personal psychology, and has >nothing to do with the meanings of expressions, which is what I am concerned >with. > The root of our disagreement, Chris, is that there is nothing "mere" about this matter. It is "the meanings of expressions" (whatever that means) that is "merely a matter of personal psychology." Such meanings are artifacts of agents which go about trying to satisfy their goals in the world. More specifically, they are side-effects of the behaviors of those agents. If you wish to focus your scholarly attention on them, that is your own business; and there is certainly much which remains to be studied. However, that does not mean that you should overlook the actual role of these artifacts in the general picture of behavior. >Finally, with regard to your last paragraph, if two people are never really >talking about the same things, then they can never come to any agreement about >the same things (since there are no such 'same' things to agree about). If you >are right (as you may be), then I marvel that people ever manage to >co-operatively get anything done in the world. The ability to marvel is healthy when it leads us to ask questions and pernicious when it just makes us stand back and gape. > It may be that people simply >behave *as though* they were talking about the same things; this opens up >quite another can of worms (what is the ontological basis for this >"as-though-ness"?) > The only reason you can use the word "another" is because those "functions from extensions to possible worlds" is an even greater can of worms! My attempt to focus on behavior, rather than all the intellectual baggage of formal semantics, is nothing more than a vigorous attempt to hurl this first can into the sea! This is not to say that we should all go out and dust off our copies of Skinner's VERBAL BEHAVIOR. We should just remember that, for any intelligent agent, the real game is one of satisfying goals. > The example I like is that of a filing cabinet. When >I use the expression "filing cabinet" as a referring expression, I intend to >pick out for my hearer something more than a metal container, around 4ft 6in >high, with a small number of moving parts. What I mean by "filing cabinet" >includes assumptions I have about its functionality, about practices of >producing and storing textual documents, about the history of such practices >within my culture, and about the graphemic storage and retrieval of >information. I do indeed have an "internal model" of a filing cabinet, and I >take it that my having such a model is a necessary condition for my being >able to use the expression as the content of a referring act. I just happen >to espouse a version of realism that allows the world to be populated with >filing cabinets (and other things like chairs) in the rich sense I outlined >above. The internal model that I have of a filing cabinet (or of a chair or >of whatever else) is derivatve and dependent on the existence of filing >cabinets of just this kind 'out there' in the world. It is by virtue of its >existing independently of any internal model that I (or anybody else, correct >or incorrect) may have of it that I can unproblematically refer to a filing >cabinet and feel confident that my hearer knows exactly what it is that I am >talking about. If my hearer gets it wrong, that's his/her problem, and I can >put him/her right. > Your example is a good one, but I am not sure you are reading it correctly. What you are really doing is justifying the necessity of dialog. It is only through dialog that you can establish whether to not your hearer has "tuned in" to your intent. Your last sentence is actually quite selfish and arrogant. The problem does not lie solely in your hearer but in both of you; and it is only through cooperative dialog that the two of you can resolve it (at least for the immediate purposes of any goals you both may have). I think I now see what you fear in solipsism. You see it as the ultimate form of that selfishness I have just criticised. I, on the other hand, see it as an invitation to a much more cooperative form of not only dialog but also other aspects of inter-personal behavior. You are, of course, free to turn down that invitation . . . as long as you do not get in the way of the rest of us! =============================================================================== Stephen W. Smoliar Institute of Systems Science National University of Singapore Heng Mui Keng Terrace, Kent Ridge SINGAPORE 0511 BITNET: ISSSSM@NUSVM "He was of Lord Essex's opinion, 'rather to go an hundred miles to speak with one wise man, than five miles to see a fair town.'"--Boswell on Johnson