Date: Sat, 4 Oct 1997 12:52:17 -0500 (EST) Message-Id: <199710041752.MAA20282@locke.ccil.org> Reply-To: Chris Bogart Sender: Lojban list From: Chris Bogart Subject: Rational thought requires ambiguity X-To: Lojban List To: John Cowan In-Reply-To: <199710040330.VAA06641@indra.com> X-Mozilla-Status: 0001 Content-Length: 1657 Lines: 30 On Fri, 3 Oct 1997, Andrew Sieber wrote: > how this particular ambiguity is allowed to remain. John Clifford > pointed out that the ambiguity can be resolved by rephrasing the phrase, > but that does nothing to resolve the issue of the ambiguity of _this > particular_ phrase, "skami pilno". I think Lojbab's comments on this topic in his recent message were very good, but I wanted to add something here: rather than attempt to eliminate ambiguity, lojban attempts to let the speaker decide exactly when and how much ambiguity to use, and to mark it as such. I think there's a compelling argument that says it's impossible to eliminate ambiguity from language: part of what lojban does, I think, is provide a more rational way of marking and managing ambiguity. As an analogy, imagine you're trying to draw the Mandelbrot set, or some other fractal picture; you have a screen of a certain size and certain resolution. When you're drawing some sub-area on the screen you use an algorithm to break the picture down further to get a value at each pixel; but when your sub-area happens to be the size of a single pixel, you want a different kind of analysis: you want to gloss over the detail within the area and pick a single color to represent the infinite complexity found within that pixel. Assuming that the world and our ideas about it are in many cases more like fractals than like traditional geometric shapes, Lojban needs tools for both precision and ambiguity at its disposal, just as English does. A tanru or a predicate with omitted places is like a pixel -- it glosses over details in order to prevent infinite recursion. co'o mi'e kris