Re: Shemspreg at FrathWiki

On 3/15/07, Henrik Theiling <theiling@...> wrote:
> Hi!
>
> Dirk Elzinga writes:
> > Hey.
> >
> > I just put the Shemspreg grammatical description on
> > http://wiki.frath.net/Shemspreg . Your comments would be welcome,
> > especially on layout and organization.
>
> Nice language.  IE is quite obvious. :-)  'wes' = 'stay overnight' :-)
Thanks! I just replied to David about the origins of Shemspreg; you
might find it interesting to see where it came from.

> Because you ask for organisation, I suggest to make the table layout
> more uniform.  E.g. some tables have number Y-axis and case on X-axis
> (personal pronouns) while others have case on Y-axis and number on
> X-axis ('this', 'that').
I'll look into that. I think that the pronouns are the only
case-inflected forms which have this arrangement, and this seems to be
the traditional arrangement for the presentation of pronouns.

> I notice that you greatly simplified the numbers, e.g. eleven=ten-one.
Yes. I toyed with keeping some of the original oddities (separate
stems for 11, 12 and 20, &c), but decided against them. If I get back
to developing the language, I may introduce more options (I do have
the option for a 'twenty' word along with a 'two-ten' word).

> Also, you simplified the case system and number&case interaction to a
> regular system and dropped person endings from verb, right?
Yes. I guess it's because I'm not used to languages with full
person/number marking or complex case systems, neither of which is
found in the Uto-Aztecan languages I work with. At least, that's my
excuse and I'm sticking with it :-).

> Was it not you who was one of those who did not like my Terkunan for
> simplifying morphology so much? :-P  You even have regular 'oinoto'
> while I kept 'prime'. :-)
Yes, well, my impression of Terkunan was that you were trying to make
a Romance language that was a historically plausible projection of
Latin. And in that spirit, I felt that your abandonment of some of the
morphological irregularities was unrealistic. My excuse is that I'm
not trying to make a historically plausible projection of PIE, and so
I should be allowed my regularities :-). But I do keep the nominal
-r/-n alternation (-r occurs word-finally, -n occurs elsewhere), which
I can't justify. So there may be room for _per(s)to_ 'first' and
_sekwent_ 'second' (< _sekw-_ 'follow') in the ordinals.

Thanks for the comments.

Dirk