[tied] Re: "Will the 'real' linguist please stand up?"

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:

> On Tue, 25 Feb 2003 10:01:20 -0000, "tgpedersen
> <tgpedersen@...>" <tgpedersen@...> wrote:
>
> >> The *-tos in the ordinals is not properly an adjective-forming

suffix

> >> (*-os is). It probably became an ordinal forming suffix by

accident

> >> (*dek^mt ~ *d(e)k^mt-os --> *dek^m. ~ *dek^m-tos)
> >
> >Erh, hmm. If Miguel states it so forcefully, it must be true.
>
> I can leave the "probably" off, if you prefer.
>
> >A quirky example from Danish: numerals fem, seks, syv, otte, ni,

ti;

> >ordinals femte, sjette, syvende, ottende, niende, tiende; cf past
> >participle -te, present participle -nde. Synchronically,

therefore,

> >these Danish ordinals look as if they were participles, thus
> >adjectives (please don't entertain me with their history, I know

it).

>
> But that makes my point, doesn't it? The suffix is historically
> -de/-te, but the dropping of -n in the cardinals makes it look as if
> the suffix is now -nde from 7 on (even though there never was an -n

in

> 8, that's analogical spread).
>

Possibly, but it makes my point too: It is possible for the
Sprachgefühl of a language to have ordinals that seem to be derived
by a adjective-forming process from the cardinals. If that is so,
some language might hit upon the idea of actually doing just that.
But it would all hinge on the existence of an adjective-forming t-
suffix in AfroAsiatic (but the prospects are slim, I think)

The other possibility to avoid your one-of-a-kind PIE t > k rule was
to assume that the PIE /k/ in "six" was loaned as /k/ from
AfroAsiatic at an early date, and Semitic later going k > t, the only
trace of /k/ left being in Omotic.

Just a possibility.

Torsten