[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Two ideas, before I forget

la mark. clsn. cusku di'e

> 1) A new cmavo in selma'o CAI (isn't it nice to see a proposal that doesn't
> involve a new selma'o?), to indicate "I'm not telling you where I stand on
> this scale."  We currently use {cu'i} for this sometimes, translating
> {pe'icu'i} as "I don't necessarily opine", but really it means "I'm between
> opining this and not."  You could argue that simply not having a UI for the
> scale would accomplish the same purpose, but it really wouldn't.  Often you
> want to note explicitly that you're not passing judgement, indicating
> approval, agreeing, etc.  Imagine a newscaster, for instance.  Or in the
> case of names made from selbri, like calling Israel {la censa gugde}.
> Using {la} does indicate that you don't necessarily really mean the selbri,
> but even so a {pe'ia'a} (for {xa'a}=my cmavo) would be nice.

I think that "ge'e", the non-specific emotion, can work this way when used
as a modifier.  Alone, it indicates that you feel some emotion that isn't on
the list, or that you aren't expressing your emotion, thus:

	A:  pei le mi mapku
	B:  ge'e

But since it makes no sense to specify "ge'e" along with a specific emotion,
I would tend to read it as a non-specific modifier:  "uige'e" means "I'm not
saying where I stand on the happiness scale".  This breaches the distinction
between UI and CAI a bit, but these selma'o are grammatically equivalent
anyway and could easily have been combined in the first place.

> 2) A fairly pointless suggestion for now, but it might come in handy:
> expand cmavo space to include [C]V'VV forms.  Thus, {*ma'oi}, {*.a'ei},
> etc.  These are no less pronouncable than anything else, and I don't see
> why they should be excluded.  'Course, we're not out of cmavo yet, so
> there's not much use for it.... (BTW, why don't I even see any cmavo of for
> xVV or xV'V?  They're not even listed as UNK (unassigned) forms!).

This has been done.  All xVV and xV'V cmavo have always been
reserved for experimental use.  (This is distinct from UNKs, which are not
used for any purpose.)  In addition, all "extended" cmavo, which involve
an optional consonant followed by three or more vowels (with apostrophes
as needed), are likewise reserved for experimental use.

So far, the only suggestion for the extended cmavo has been to use them
in machine translations to express the different flavors of tanru: thus,
"blanu goi'o zdani" might mean "house which is blue", whereas "blanu
gai'o zdani" might mean "house for blue inhabitants".  The richness of the
tanru concept means that many, many such markers would be needed.

Current machine parsers treat all UNK and EXP cmavo as belonging to UI,
simply because that makes a successful parse more probable.  "ci'a", e.g.
is currently UNK, but most texts using it will still parse even if

John Cowan	cowan@snark.thyrsus.com		...!uunet!cbmvax!snark!cowan
			e'osai ko sarji la lojban.