[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
--- In ceqli@yahoogroups.com, MorphemeAddict@w... wrote: > In a message dated 8/9/2005 10:44:20 AM Eastern Standard Time, rmay@m... > writes: > > > > > > > While you are the first to mention the name, and I do keep up with the > language, I've regretted choosing that name for a long time. > > I'd really appreciate your input into the whole name problem. > > > > > Lojban is like Loglan in *most* respects. Only the words were changed to > protect the guilty. (Basically). > > I'm wondering about names, mainly. > > I know that JCB decided that one o'clock was a name, for some reason, and > > I've often > > wondered if that makes any sense at all. One thing I'm pretty sure of is > > that one o'clock, > > Monday, and January are _not_ names in any useful sense. > > > > What useful sense did you have in mind? One names things that are unique (or > nearly so), and one o'clock, Monday, and January all are unique. There is > only one of each in any given larger unit. (In fact, I'm surprised no one has > named the weeks.) An even weirder name is "1995", for example. But it too is > unique. Consider how they're used in English: They do not take the definite > article "the" (except under unusual circumstances). "In January" vs. *"In the > January", or "January is a long month" vs. *"The January is a long month". > > > Another problem is that Canis Familiarus seems to be a name, but kanin > > doesn't. > > > I agree. I'm not sure why, though. > > Or am I > > just trying too hard to systematize something unsystematizable? > > > > Not at all. You're trying to make sense out of chaos. Everything is > 'systematizable'. But some collections have more order, more system, than others. > > I recommend (all of but specifically here) Chapter 24.0 on names in Rick > Morneau's Lexical Semantics: > http://www.eskimo.com/%7Eram/lexical_semantics.html > Thanks. I bookmarked it. Backing up to Lojgblan, I found the initial idea of word categories very stimulating. Little words, predicates, and names. Trouble is, that the little words turned out to be a real catch-all. So I sort of forgot about all of that in doing Ceqli. As I said elsewhere, numbers as little words just didn't feel right, but to suit L grammar, they had to be classified as such, evidently. So, I think there are limits to systematization. It's easy to say that janzo is a name, and everybody agrees. For the situations in which you have two Johns, you can always say "dusi jin hu kiam toi janzo betoi". In normal usage, you'd probably just say "du janzo." But it gets hairy when you get into long, complex names. I frankly don't know what to do with 'the National Association of Conlang Quibblers,' because the parts of the name have lexical meaning that you don't want to lose, but adding -zo to make "janzo" is specifically for the purpose of clarifying that it's _not_ the same as "jan." As "japanzoheim" isn't really go-bread-land. Suppose I just say that any compound ending with say, -heim, is a name. Then I have "japanheim". Fine. But then "gerheim" (foreign land) would have to be a proper name, which it isn't intended to be. So to go that way, I'd need a unique compounding form for any category a name might represent ? country, language, nation, ethic group, river, pond, mountain, planet, galaxy, etc., probably infinite. Speaking of infinity, names are of course potentially infinite. As it stands, "japanzo" is a vague word meaning "Japanese" that just about has to have some other word combined with it to make it meaningful. "japanzobol/heim/jin". Come to think of it, we could think of -zo as a _prefix_ of sorts on the following word ? heim, whatever ? which turns it into a morpheme that makes what precedes a proper name. Or the whole compound, actually. I suppose I need a t-word for words like "The Democratic Party" after all. (Analagous to Loglan "la"). I know Loglan had a very clunky way of forming predicates meaning 'french' from the proper name of the country, which is going way too far for me. "franzojin", "franzobol" and "franzo sa" seems to do the job there. Sometimes I think the way to go is to completely forget about the name/morpheme distinction in general, keep -zo for the purpose of converting any morph or compound into something _without_ lexical meaning, therefore having no possible use except as a name, and letting all other names run wild. Then it would work like "Vulcan". You know ? "I'm a Vulcan from Vulcan and I speak Vulcan." "Go bi hamerizo hu de hamerizo kai go bol hamerizo." Expanded to, if necessary: "Go bi hamerizo jin hu de hamerizo heim kai go bol hamerizo bol." Probably, in practice, without the spaces: "Go bi hamerizojin hu de hamerizoheim kai go bol hamerizobol." And dealing with longer things, "to di sa ze fren" would be "the friends of animals," and "ti di sa fren" would be "The Friends of Animals" as a name. Unless it would be allowable to append -zo to the to and "tozo" would serve as the "ti" That's worth thinking about. Save a t-word. I've thought about it awhile, and I like "tozo". To me, January 1st, 2005 is like "the tallest man who ever lived" ? it's unique, but it's not a name. Trouble is, I don't think Wednesday is a name but I do think Christmas is. Why, I can't say.