RE: [jboske] unresolved debates

["And Rosta" <a.rosta@lycos.co.uk>]

John:
> And Rosta scripsit:
>
> > So basically they mean pretty much the same as their English
> > glosses. As opposed to collectively forming some sort of
> > 5-point scale
>
> I wish to go on record (as the virtual pc, it seems) as absolutely
> disagreeing with this. The so'V are and were meant to be just that,
> a five-point scale; the English keywords are imprecise, and deductions
> from them are illegitimate. So saith Woldy and so say I

[Is that last sentence an echo of a refrain in a Villon poem? It
is nigglingly ringing a bell I can't identify.]

This is clearly spelt out pp440-441.

Let me solicit comments on the following points, so I can compose a
summary for the wiki.

In favour of the "English keywords define the meanings (in this
instance)" position:

* It's more useful. (cf natlangs)
* Usage overwhelmingly supports it; pretty much all prior usage
of so'V would be invalidated.
* Other mahoste glosses, such as {so'e roi} = "usually" seem to
support it.
* We would need to find other ways to say "many/a few/most".
* It's hard to believe that the keywords were so incompetently
chosen. Why wasn't {so'u} "almost none", and {so'i} "a middling
number" or "about half", or suchlike?

In favour of the Woldyan position:
* Thus is it Written.
* It is sort-of consistent with the phonological patterning: they
all form a so'V series, whose ordering makes sense if you take
it to be alphabetical so'a>so'e>so'i>so'o>so'u counting in
the opposition direction (big to small) from normal (small to big).
This has the virtue of accounting for the apparent antiiconicity of so'i/so'u.
(I'd have expected so'u>so'o>so'a>so'e>so'i, myself,
though; much less arbitrary than alphabetical ordering.)

--And.