2011/7/20 Patrick Dunn <[log in to unmask]>:
> 2011/7/19 René Uittenbogaard <[log in to unmask]>
>> For Verb-final and Subject-final languages, this does not seem to
>> far-fetched.
> That morpheme *only* occurs word finally as that morpheme?

Ehm, no; sorry if that was unclear. Those phonemes also occur word-initially.

2011/7/20 Alex Fink <[log in to unmask]>:
> On Tue, 19 Jul 2011 20:55:51 -0500, Patrick Dunn <[log in to unmask]> wrote:
>>2011/7/19 René Uittenbogaard <[log in to unmask]>
>>> 3. How plausible would it be to have this mood marker on any word
>>> that happens to be sentence-final?
>>Pretty darn implausible, sad to say.  Now watch someone say "well, this
>>natlang here does it."  :)  But unless there's a whole slew of such Natlangs
>>that do, I'm willing to say -- unlikely.
> I think it's clear sailing if you call the mood marker not a suffix, but a
> clitic.  Syntactically let it be just the last word of the clause;
> phonologically bind it to the previous word.
> If this were my lang, I'd be tempted to make the full forms of the mood
> clitics CV and have a prepausal vowel deletion rule (a common thing).  But
> that would introduce at least two complications -- what happens if that
> leaves an obstruent as final C?  what happens when it applies somewhere
> there isn't a mood marker?  So you'd do fine not to do this.

Making it into a mood marker and cliticizing it looks like a good idea.
The mood particle could move to a different place if the flexibility is needed,
and I suppose an obstruent final C as a clitic to a CV-word would not be
a problem at all. And we won't be needing plosives in that position.

> Honestly, what I dò find unlikeliest about the scheme, or at least the
> biggest tell of being a conlang (if the presentation weren't framed in this
> "punctuation" style, at least), would be the too-close correspondence
> between the set of resonants and the set of modes in the language (unless
> you vary it up some).  Conlangs are always doing this, making the set of
> forms possible in some morphosyntactic paradigm fill out a very large
> portion of some natural class of phonological sequences; natlangs are rarely
> so lucky.

Thanks for pointing this out; however, I think my conlangs still have
quite a few
properties that give them away as conlangs, so I'm already happy if those
properties aren't "pretty darn implausible".