Re: First Post and Proto-Conlang rough sketch

> Is there a design goal behind excluding 'r' when 'l' is allowed after
> plosives?  It would feel like a natural pair to be included, with the
> possible exception of excluding 'l' after alveolar stops.  (This may
> be a Latin-centric question :-)).
>
> Similarly, since the 'w' glide is allowed, is there a reason to
> exclude 'j'?
> This is a question out of curiosity -- I could imagine reasons or
> design goals easily that would make the above the first choice.
Yes, there is definitely a design goal:

I want the roots to be "re-vowel-able(?)" from their zero-grades, and if I allow
the syllabants on either side of the vowel, then ambiguity arises.

kwes > kws [kus]
kews > kws [kus]

kws > which one? kwes or kews?

The reason they must be able to be re-voweled, is that other words
can "donate" their stress to it, i.e. giving back the zero-grade its e.

I happen to like labialized plosives (pw, tw, kw, etc . . .) and I found that r
and w seem to easily mix (as in some dialects of English [or kiddie
pronunciation?]) so allowing, for example, kl, kr, and kw could, in theory, give
rise to kl, kw, kw . . . and while that allows great potential for a future
daughter lang, its not particularly a direction I'd LIKE to go in.

At this stage, I want very little ambiguity. Later, as sound changes are applied
plenty will arise, but I would like to be able to over see the
analogical "smoothing-out" processes that typically follow the appearance of
ambiguities.