search

lojban / cll

Complete Lojban Language Chunked
http://lojban.org/publications/cll/
Other
176 stars 49 forks source link

Equivalence of stage-3 fu'ivla with different rafsi forms #446

Open Wuzzy2 opened 3 years ago

Wuzzy2 commented 3 years ago

The CLL states that 2 lujvo are considered equivalent if all their rafsi refer to the same “parent words”.

However, the CLL doesn't say anything similar about stage-3 fu'ivla. This is a question I'd like to see clarified in the CLL:

Are two stage-3 fu'ivla that only differ in their rafsi that both refer to the same “parent word” considered be equivalent in meaning (similar to lujvo equivalence)?

E.g. is “gism,r,bla” considered to be equivalent to “gim,r,bla” in meaning, or are those two completely separate words with (possibly) different meanings?

phma commented 3 years ago

I think they should be considered alternative forms of the same word, just as "xauzma" and "xagmau" are alternative forms of the same word.

Wuzzy2 commented 3 years ago

I agree.

vpbroman commented 3 years ago

Defining these equivalences, do we then have to impose restrictions on stage-4 fuhivla that they must not look like a stage-3 fuhivla? If stage-4 forms may collide with stage-3 forms, we have to know the etymology of a fuhivla to be sure what forms might be equivalent to it and how to look it up in a lexicon.

Wuzzy2 commented 3 years ago

I don't think so. A stage-3 fu'ivla is clearly defined to be of the form RAFSI + HYPHEN + REST.

Like cid,r,spageti. The fu'ivla stage-3 is, as far I understood, only defined on a grammatical level. So if the fu'ivla matches this syntax, it is stage-3, otherwise it is not.

The notion of a "stage-4 fu'ivla that looks like a stage-3 fu'ivla" doesn't make sense to me.

So no. No restrictions on stage-4 fu'ivla need to be applied, it wouldn't make sense.

Anyway, this is besides the point.

I noticed something. Currently, the CLL seems to state that two stage-3 fu'ivla with different rafsi forms (but for the same word) are "distinct", as suggeted here, in the fu'ivla chapter:

Technically, “ricrxaceru” and “tricrxaceru” are distinct fu'ivla, but they would surely be given the same meanings if both happened to be in use.

It is a bit unfortunate that the CLL didn't add an equivalence rule like for lujvo. It might be too late to change that now …