Yesterday, we talked about intelligence and truth. One of the points that was made was that intelligence is basically applied knowledge -- that is, it doesn't help someone if he knows a lot but doesn't apply this knowledge in his life. We also talked about how truth basically is the knowledge of unchanging principles.
An interesting parallel was drawn between knowing a gospel principle and knowing a secular principle such as grammar or mathematics -- if someone knows a gospel principle and applies it in his life, then that is often termed "obedience". Yet the action is basically the same as when someone takes a grammatical rule and uses it to string along words to make a sentence, or someone who takes two mathematical laws and uses the principles of logic to derive a third law from the two of them. In both cases, there is knowledge of a truth, and the application of that knowledge. So in a way, a person who writes a grammatical sentence is obeying a rule or an underlying principle in the same way that someone who observes a gospel principle is doing so -- or an architect who obeys certain rules in designing a structure, or (I suppose) a programmer who implements an algorithm.
Re:principle is not necessarily truth
pne on 2002-03-25T08:26:24
Well, since this discussion was in the context of a religious class, it was posited that there *is* an absolute truth -- but that all gospels only understand part of it (a bit like the elephant and blind men story, perhaps): some understand, say 80% correctly, others 55%, others perhaps 92%, with overlaps and so on. This would imply that if two gospels have different views on the same subject, they could not both be completely correct at the same time.
(And about not fighting a war over mathematical truths, wasn't Galileo or someone killed for refusing to believe the earth was flat?)
Instead of supposing that generating sentences works by following rules for assembling the parts, the functionalists would say that there's instead a bunch of neural networks that say how much a given sentence looks like ones it's heard before; so [the functionalists argue] actually producing a good sentence is a matter of applying a genetic algorithm that starts out with a rough stab at a sentence, mutating it, running the mutations past the "how familiar does that look?" neural networks, and then picking the one that looks the least weird.
In that model, any actual generalizations like "pronouns should have a clear antecedent" are not the operational principles of the system, but merely generalizations that we would make about the behavior of the system.
This approach's application to math, morality, ethics, or even religion, is left as an exercise for the reader.
Re:Grammatical "truth"
pudge on 2002-03-08T18:25:47
In that model, any actual generalizations like "pronouns should have a clear antecedent" are not the operational principles of the system, but merely generalizations that we would make about the behavior of the system.
While I might agree in principle, this to me seems like a bad example, since without a clear antecedent the pronoun is vague; so this is not merely a matter of proper form, but of adequate function... Re:Grammatical "truth"
TorgoX on 2002-03-08T20:29:01
Case one:
Say you and the wife are at a restaurant. A guy comes in and the waiter leads them to the empty table next to you. The guy takes his jacket off and puts it on the back of the chair. He picks it up, and in doing so, somehow spills the water glass all over himself. You and your wife both see this. As the guy goes off to the bathroom to towel off his clothes, you look at your wife and say "wow,/he's/ having a bad day!". Voila, a pronoun with no antecedent -- but a clear referent. Case two:
In some languages, nominalization can do odd things to pronoun reference. So "she brings us food" is a perfectly good declarative sentence in such a language, but add a nominalizer suffix and "she" stops requiring an antecedent, via a construction which in this case I could translate as "she who brings us food" [altho English requires the relative clause there, because it has no clausal nominalizer suffix]Case three: "He who laughs last, laughs best!" "He" doesn't have an antecedent.
Re:Grammatical "truth"
pudge on 2002-03-08T22:56:29
I guess when I was thinking "clear antecedent" I was thinking of referents as well; that is, that in case one, we clearly know what the referent of the pronoun is. Me use word bad.
ly.
