On Sun, Aug 16, 2009 at 1:44 AM, Pierre Abbat
<phma@phma.optus.nu> wrote:
On Sunday 16 August 2009 01:02:30 Luke Bergen wrote:
> what's this about a revision of phonotactics and morphology? Have these
> changed since the CLL?
There's a proposal to switch from the LALR grammar to a PEG, and there are
some points of morphology on which the CLL is inconsistent, confusing, or
underspecified. Nothing is going to change with gismu or lujvo morphology
(except possibly allowing "y" where it's not needed). Only some corner cases
of fu'ivla and cmevla may be affected. Examples:
* One word in the Book (i,iai,i,iai,ion) has a triphthong. There is no list of
allowed triphthongs. If a triphthong is allowed to be the nucleus of a
syllable in a lujvo, then "mliau" (meow) is one syllable, therefore invalid,
and "amliau" is valid. If there are no triphthongs, then "mliau" is valid,
and "amliau" breaks up. Either way, "mliaue" is valid.
* The fu'irvlazba algorithm, applied to "jamo" with the three-letter rafsi,
yields "lerndjamo", which is invalid. It has to be "lerldjamo".
* The Book says that a cmevla can't contain "la", "lai", or "doi", unless
preceded by a consonant. This doesn't make sense: "lai" contains "la",
and "la'i" isn't mentioned. My amendment is that a cmevla can't
contain "la", "lai", "la'i", or "doi" unless preceded by a consonant or
followed by a vowel or apostrophe. This means "laus" and "ala'um" are both
valid. The dot side throws out the rule entirely, and puts pauses before all
cmevla.
Enough of this! It's too much for the beginners' list.
Pierre