[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Lojbab: > At 06:09 PM 1/13/03 +0000, And Rosta wrote: > >This is why everybody hates zi'o. It was proposed as a fix to > >blotation (Bloated Gismu Syndrome, where the basic gismu has > >places that shouldn't be there and should have been addable by > >BAI or lujvo) > > Whereas I wanted such to be solved by lujvo, but there was no jvajvo rule > that provided for deleting a place from a bloated gismu and people seemed > unwilling to use non-jvajvo means or to define a zil without having a > corresponding zi'o or to come up with some sort of other rule on how to do it > > >So that you can, say, talk about bottles in > >general rather than lidded bottles in particular. Or tigers > >in general, rather than tigers with stripes > > Whereas by Lojbab's founder-intent (as opposed to Cowan's) an unlidded > bottle is simply not expressed with botpi - zi'o or otherwise, but as some > other kind of vasru. Just like a "nu zi'o catra" is not expressed with > catra but with mrobi'o. Thus the Lojban words simply map semantic space > differently from English (or arbitrary other natlang), rather than using > heavy-handed techniques to make Lojban words work in the same way English > words do I wager that if you consult the languages of the world you will not find one that has a basic vocable for 'striped tiger' and no basic vocable for 'tiger'. The clamour for some device to get rid of sumti places was really focused on the ones that should never have been there in the first place. It is not enough to say that Lojban rather arbitrarily maps semantic space differently from natural languages, since natural languages' basic vocab will reflect cognitive universals of basicness. However, it turns out that zi'o is rather a good solution after all, in the circumstances (viz that gismu place structures are frozen), as I've written about in another message. > >But it's incredibly counterintuitive to mark "there may or > >may not be a lid" and "there may or may not be stripes" by > >an overt word. When you think of bottles and tigers you > >don't normally stop to check whether or not you are thinking > >only of lidded and striped ones > > Then you aren't thinking of botpi and tirxu, but vasru (possibly patxu) and > mlatu I know I'm not thinking of botpi and tirxu. I'm thinking of bottles and tigers, of bottiglie and tigri, of animals of species Panthera tigris. > If this means that I'm inclined to think of a zebra as more of a tirxu than > a stripeless tiger is, well, that's how the place structures work > > >And it's all to easy to > >intuitively misinterpret zi'o as meaning "is a lidless bottle", > >"is a stripeless tiger", in distinction to noda, "is not > >a lidded bottle", "is not a striped tiger" > > > >We all hated zi'o right from the start, and it is really only > >there in order to shush the people complaining about blotation > >As a practicable solution to blotation it fails, which is why > >I advocate that elliptized zo'e should mean "zo'e a zi'o" > > But if "zi'o" is effectively a superset of "zo'e" then "zo'e a zi'o" means > merely "zi'o" Allowing that zo'e refers to a specific phrase, which is to be glorked, gives more satisfactory results. --And.