Wiktionary talk:About Lojban

Parts of speech
Since this is the English wiktionary, POS headers must be in English. (Not meaning that they are restricted to the eight English POS, but that they must be in English.)

Is simply a matter of using the correct English term for the POS. "gismu" is Root, "rafsi" is Affix, "cmavo" is Particle, and so on. (note selbri doesn't even have an entry!) New English headers may or may not be needed.

It would also be nice if this page was written in English, it is nearly incomprehensible. Robert Ullmann 13:37, 22 November 2007 (UTC)


 * English-language works on Lojban use the terms gismu (for which I think "root" would be inaccurate anyway), rafsi, cmavo and so forth. Are we stop using terms like hanzi and romaji for Chinese and Japanese as well? --Ptcamn 16:56, 22 November 2007 (UTC)


 * "hanzi" and "romaji" are both accepted borrowings into English (millions of Google, g.b.c). "gismu" is an obscure word from a minor language that simply isn't English. Robert Ullmann 09:44, 23 November 2007 (UTC)
 * For example from Random House Unabridged. Robert Ullmann 10:38, 23 November 2007 (UTC)
 * Also it should be noted that we do not use "Romaji" as a POS header for Japanese (it still appears in a handful of entries, but is severely deprecated). "Hanzi" is used as a non-POS L3 header only for single characters. In no case whatever is a non-English or borrowed term acceptable as a POS header. Robert Ullmann 13:12, 23 November 2007 (UTC)


 * Syntactically a brivla is a verb, but semantically it covers verbs, common nouns, adjectives, and adverbs; calling it a "verb" is inaccurate. Ditto on hanzi and romaji; also Hebrew grammars use words like "lamed he" and "niph`al" in the middle of English sentences. PierreAbbat 23:41, 22 November 2007 (UTC)


 * And we wouldn't accept "lamed he" in a header either. Robert Ullmann 09:44, 23 November 2007 (UTC)


 * Our definition of brivla says it is a predicate. So is that wrong? If yes, the definition should be fixed. If correct, then the header can be Predicate. Or whatever the English definition of "brivla" is. Robert Ullmann 10:07, 23 November 2007 (UTC)


 * "hanzi" and "romaji" get millions of Google hits because they are used to describe languages spoken by millions of people. "gismu" is an obscure word because Lojban is an obscure language.
 * "simply isn't English" is a meaningless accusation. I could just as easily accuse "romaji" of really being Japanese and not English. "gismu" is, as I said above, used in English-language works on Lojban. If that doesn't make it an English word, nothing does. --Ptcamn 19:55, 23 November 2007 (UTC)


 * Connel once suggested that the part of speech for a brivla should read "verb/noun/adjective/adverb". I don't know. Incidentally, some particles are also brivla, so I don't know which part of speech you would use in that case.&mdash;Nat Krause 22:11, 23 November 2007 (UTC)


 * I think "predicate" is better than "verb/noun/adjective/adverb". The latter implies that the word belongs to four parts of speech; it just belongs to one.
 * By particles which are also brivla, do you mean go'a and the like? Those are the ones I meant by "pro-verb". PierreAbbat 02:05, 24 November 2007 (UTC)


 * "Predicate" is too encompassing. All brivla are predicates, but not all predicates are brivla (e.g. "pa moi" or "pu klama" are predicates). "Predicate word", which is where "brivla" comes from after all, is more accurate. Xorxes 15:12, 27 November 2007 (UTC)


 * Changed to "Predicate word". What about pro-bridi? Is there a better term for this?

Definitions
The definitions must be actual definitions (translations to English). "x1 something x2 something x3" is not a definition.

(It might be useful to use an L4 Construction header, as in WT:AJA. Or might not.)

ke is a reasonable example of an entry in English for a Lojban particle. Robert Ullmann 13:37, 22 November 2007 (UTC)


 * ke is a particle; phrases of the form "x1 something x2 something x3" occur only in definitions of brivla. To translate mitre simply as "meter" is wrong; it would lead people to translate "three meters" as ci mitre, which actually means "three things each of which is some number of meters long". PierreAbbat 23:41, 22 November 2007 (UTC)


 * That's why you have examples and usage notes. To explain the language in English. You know, perhaps Lojban should simply be removed from CFI, if no-one can properly describe it in English. Robert Ullmann 09:46, 23 November 2007 (UTC)


 * You seem to be getting confused: the "explain the language in English" argument is the one above this one. This one is about the format of entries, nothing to do with the language used.
 * Many Lojban gismu take several arguments whose meaning isn't obvious. Por ejemplo:
 * la .iulias. klama la .uacintyn. la losandjeles. la cikagos. la .amtrak.
 * Julia travels to Washington from Los Angeles via Chicago on Amtrak.
 * (lit. "Julia go Washington Los Angeles Chicago Amtrak."; to, from, via and on are contained within the meaning of klama, which could be described "tetratransitive")
 * It's only sensible to give these meanings as part of the definition. To not would be like defining the English word kill as "die", with a usage note saying "The one that dies is the object, and the one that caused the death is the subject." --Ptcamn 19:55, 23 November 2007 (UTC)
 * The Construction header looks like it might be reasonable. Could you write an example? PierreAbbat 02:05, 24 November 2007 (UTC)

Template:rafsi of
Please comment: Template talk:rafsi of. --Μετάknowledge discuss/deeds 07:05, 9 September 2012 (UTC)