Hallo!

On Mon, 18 Apr 2011 20:00:14 -0700, J. Burke wrote:

> >I see.  But the prefixes must have come from *somewhere*, and it
> >strikes me as weird to assume that they flipped over from one end
> >of the verb to the other.  They must have been either independent
> >pronouns, or suffixed to something preceding the verb which got
> >eroded away.
> 
> The pronominal affixes didn't "flip" from one end of the verb to the other, 
> if you mean that literally; the old and new affixes are actually distinct 
> entities, though the latter derive from the (ruins of the) former.  The 
> _idea_ 
> of where person should be marked on the verb is what really flipped.  

Sure.

> Whether the 
> new prefixes were originally independent pronoun words or not is kind of 
> beside the point, and not really of interest to me; I know that the location 
> of
> person marking on verbs can change in that fashion without leaving a trace 
> of
> any intermediate developments, and, frankly, no one knows why or how it 
> happened 
> Algic.  I'm content to leave it at that, since there must be an explanation 
> for 
> it, and because we just don't know.  There are tons of issues to work out
> for any conlang, and I've preferred to focus on other matters; if, in the
> future, someone figures out the Algic pronominal flip, I may go back and
> devise a suitable explanation for Hanoa?tsi based on that.

I see.  That is of course legitimate.  If Algic did it, your conlang
is of course free to do that as well.

> >> They served to mark an intransitive participant as a patient/undergoer, 
> >> in 
> >> the context of the active agreement system.  -š- was used in inanimate 
> >> third 
> >> person forms, while -ł- was used for all other persons (both ultimately 
> >> descend from the same element, the pre-Proto-Central Mountain root *-tɛ- 
> >> 'accidentally'; by various twists and turns, they developed into those 
> >> different allomorphic forms in different verb paradigms, and when in 
> >> Noyahtowa the inflectional endings underwent heavy leveling,
> >> -š- happened to be the form that was present most commonly among the 
> >> inanimates, while -ł- was the most common form elsewhere; and the 
> >> allomorphs 
> >> got carried over into Hanoa?tsi when its pronominal system was reformed,
> >> as described in my first post).
> >
> >As I don't know the details of your conlangs, I cannot really say
> >whether it makes sense or not.
> 
> The semantics make sense--accidentalness > lack of initiation/control (which
> is what the affix signals on verbs).  

I see.

> >But aren't there two different suffixes of the shape -ali involved
> >here?  I mean, "obviative singular" is not the same meaning as
> >"inanimate plural".
> 
> Both suffixes are identical, -ali; you can say there are really two
> different morphemes that just happen to have the same shape, but
> that still won't help with the ambiguousness of it; which is why, in
> one case, third person -(w)a- precedes -ali- and in another follows
> it.

This is actually nothing I have problems with.  Again, I have
to say that I know almost nothing about the natlangs you are
drawing inspiration from, and cannot say what really happened
there.

> >I still find such a switcheroo rather hard to believe.
> 
> I actually had something rather like what Alex suggested in mind already:
> only singular and plural existed originally, but with free variation in 
> person-number order, 
> followed by fixing of person-number order, with one set of forms remaining 
> as plurals 
> and the variants becoming re-analyzed as duals.  But he helped me solidify 
> the details.  
> The addition of a third grammatical number would have been helpful for 
> disambiguating 
> 3rd persons, given the prior loss of the obviative.

I cannot say that this cannot work out, but it still strikes me as
odd.  But I know that natlangs can be very, very odd, and there may
be a natlang precedent for the weirdest things in conlangs.  After
all, it is *your* conlang, not mine, so do what pleases you ;)

> >Sure, you wanted to do "something different" here, but this "something
> >different" feels *very* unnatural to me.
> 
> Why so?

I can't put my finger on it, but it just feels "wrong" to me.  So
far, I haven't seen anything like that in any language, but that is
of course nothing else than a European (especially Indo-European)
bias.

> >I don't know the relevant facts here, but it still sounds strange
> >to me.
> 
> Why is it strange to omit a redundant element?  It's not even strange
> to omit a non-redundant one; e.g., non-masculine (feminine, neuter,  
> indefinite) 
> genders all fall together in the non-singular by omission of elements
> that could distinguish them.  And the vast majority of languages don't
> have an inflection for the singular number; it doesn't exist (save in 
> Bantu) because it's not needed.  Personal affixes and the like tend to
> follow patterns in their construction, but they're not literally
> mathematical-like equations where you simply plug elements into fixed
> slots, with a slot for everything and everything in its slot.  They're
> messier than that; sometimes this is due to sound change and merger, but 
> just as often it's been that way from the beginning of the system.  If
> anything, _not_ having oddities and omissions like the above would be 
> decidedly
> unnatural.

Yes, you are right.  It is not really strange to omit a redundant
element.  Weirder things happen in natlangs.

> >Perhaps I am simply too entrenched in Indo-European patterns and know
> >too little about indigenous languages of North America, but your
> >diachronics feel quite unnatural to me.  (I also noticed how you did,
> >in your diachronic developments of Hlohlamelo and Kapakwomak (if I
> >spelled them correctly) from Proto-Central-Mountain, have the same
> >few sound changes happen repeatedly, which rubbed me as wrong.
> >I have also seen your PCM pronouns, which struck me as implausibly
> >long-winded.)
> 
> They're unnatural in what way?  And which changes are unnatural?

The individual changes are not unnatural, but I had the impression
(maybe because I did not read attentively enough, I had to admit)
that the same few things happened all over again and again.  What
regards the long-winded pronouns, I was indeed referring to the
emphatic ones.  The personal endings are OK.

To conclude: it is your conlang, and you are free to do what feels
right to you.

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html