From lojbab Thu Aug 8 00:09:03 1991 Return-Path: Message-Id: Date: Thu, 8 Aug 91 00:08 EDT From: lojbab (Bob LeChevalier) To: lojban-list Subject: response to Dave Matuszek Status: RO I'll try to do this quickly from the comments as they flashed by, so my apologies Dave if I misunderstood something from my quick read. 1. I was stating the argument from the linguist's point of view, which I have admittedly pretty much adopted. But recognize that what I said, even if it is a turnoff, is reality. Linguists will not be much interested unless we can meet THEIR standards. A Sapir-Whorf test is a linguistics experiment, and linguists have to be involved and satisfied that their involvemnet is meaningful. Now you can say that their definition of language is too limited. But they are basically saying that FORTRAN is indeed NOT a human language, does not reflect human linguistic behavior, but rather computer behavior. They are interested in the cognitive processes that go on in our heads in processing language, and the assumption is that processing a code is NOT the same as processing a language. You analyze a code, you understand a language. I do not say that Esperanto is a code - I said specifically that it is an early creole. A language stops being a code when it becomes a developed pidgin. But a pidgin still has a community of speakers (and listeners). 2. Which points out that the spekaer/listener interaction is not orthogonal to my claim. The moment you have speakers and listeners you have the start of a community. 3. Yes you can say that all meanings of symbols in a symbol system is arbitrary. But in a code, the assignments are apriori arbitrary. In a language the meaning of the words is inferred by the listewner, and the fact that such inference is even possible means that FOR THE LISTENER the meanings are no l longer arbitrary, but are guessable from some experience in using the language and the specific context. In a code, you don'e interpet by guessing the meanings (unless you are a cryptographer), you just substitute from your knowledge of the apriori schema. In a human language you don;t know the entire schema so you have to make an intelligent guess. It is this use of intelligence to understand that makes atural language processing meaningfully difficult. 4. I am not hostile to General Semantics, though I admit to knowing not that much about it. It is merely one theory among many, and I am doing my best to make the language design independent of various metaphysical theories, and especially those that linguists are prone to dismiss. If the language is going to ever have practical application (i.e. if it is ever to be a real langauge), we have to sell the language to one of the three 1) an international language community 2) the artificial intelligence mnatural language processing community 3) the linguistic communtiy. 1) is downplayed so as to avoid competing with Esperanto. 2&3 are highly interdependent in that many NLP people are trained as linguists. (There are some lesser possibilities, like using Lojban as an education tool, that could gain us practical application. But almost all applications for the language require at least some degree of acceptance by linguists.) 5. Another way to look at the code/language distinction is the difference between objective and subjective meaning. In languages, we determine meanings subjectively because we cannot determine them objectively. And I am claiming that if we could do so, we would not have a language. You can have lots of clues to aid a person in learning the subjective process, like our unambiguous resolution of rafsi in lujvo compounds, but the process is still subjective because each of us has differing experiences on which we base our understandings of meaning. Recognizing that your listeners will undertsnd your words differently than you] do as you speak them is basic to language use, although probably totally subconscious when we speak our native language. It is, however obvious when we learn a foreign language, that the natives we speak to interpret the words somewhat differently than we express them. But if we talk in a code - say pig latin - to be rudimentary, we do not undersatnd the words as having meanings in themselves - we do not "attribute" meanings to words. Instead we just assign the English equivalent of each pig-laitin word, and then 'undersatnd' the English. What I am claiming is that it takes a community before you have the critical mass necessary to have words mean NOT what you want them to mean, but rather what the communal thinking has evolved them to mean, at which point you are a recipient of language, not an inventor of it. lojbba