Re: [lojban] Re: Possible polysemy clashes for "regular affixes" in lujvo?

2015-05-12 20:51 GMT+03:00 iesk <pa.fae@gmx.de>:

This is an interesting observation. However, there already is an established way to deal with such apparent conflicts of (possibly) meanings: the correct meaning is defined by the authoritative dictionary.

The problem is deeper. There are jvajvo, naljvajvo and cimjvo (the distinction between the last too is vague). It's pretty straightforward to just declare all of lujvo naljvajvo/cimjvo but then the whole idea of lujvo is ruined. What's the purpose in them if they aren't automatically expandable into bridi? Mnemonics? But then there is no need in mapping rafsi to gismu. This is in fact what it was back in 1960 with Loglan.

Then {pensku} could mean {pensycusku}, not {penmycusku}.

However, usage has established some patterns with using rafsi that bring them to the level of slotting, i.e. a rafsi take a certain position with a certain function and meaning (it is theoretically possible that in different position the same rafsi would have slightly different functions and expansions but I can't recall anything similar).

For example, {x1 rodgau x2 [x3]} is probably always expanded into {x1 gasnu lo su'u x2 broda [x3]}. At one point there was even a project (a successful one imo) that listed most used filjvo.

Another usage pattern is with -sku lujvo. They are presumed to be naljvajvo:

smusku, retsku, spusku, piksku unexpectedly break their place structure and stick to the one from {cusku}:

retsku => cusku lo preti

piksku => cusku lo pinka

and that's about that. The other lujvo in this series don't follow this pattern.

However, we have the pattern, we know what place structure to expect. This is much more than just mnemonic hints of rafsi.

I don't know exactly how zan- and mal- work. I have a feeling that they work more like {zankem-} and {malkem-} with the expansion:

lo zanbroda => lo broda noi zabna

Another pattern is -pre, which can be (may be not) a calque of Standard Average European suffix for job.

(ladrypre = milkman?).

Obviously, these example are just islands in the Sea of Cimjvo Instability (and inpredictability).

Nevertheless, they have to be described since mnemonics of rafsi helps remember lujvo to a very small degree. The most important aspect of any brivla is how its place structure parallels other words the speaker is already familiar with.

Tanru are coined on the fly. Lujvo mean what they mean. Cf. CLL 4.5, CLL 4.10.

So maybe the perception that there is a problem with lujvo stems from a slight confusion of their function in the language?

One solution I thought of might be to ditch the school of thought that all lujvo that have the same ve lujvo are equivalent and completely interchangeable
Thus {zangasnu} could be {zabna fa lo nu X1 gasnu X2},
While {zabnygau} could be {X1 gasnu lo nu X2 zabna X3 X4}

Kontraŭ-Fundamenta, thus bad. (I believe it to be one of the unnaturalistic traits of the language that word-forms as different as {xamgyzmadu}, {xaurmau}, and {xagzma} may never acquire different nuances of meaning. But that's how it is.)

{ta'ocu'i pe'i} that Lojban as-is relies way too much on the short rafsi anyway, which may inhibit learning it naturally. For example, {snura} "to be safe/secure", often gets combined with {gasnu} to make {nurgau} "to secure/protect". But that relies on the student remembering the rafsi of a not-quite-so-promiscuous gismu. It would probably be much easier to see the connection if the "standard form" remains {snurygau}.

You need to *know* the lujvo's meaning anyway (see above); the rafsi are just kind of memory hooks. And would it be wise to tweak the language around learners' difficulties in remembering parts of the language?

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.